Compare commits

...
Author SHA1 Message Date
CreatixChu 0a15e36e7f Merge pull request #4171 from deepseek-harness/worktree/release-dsh-0.1.6-alpha.1
release(dsh): 0.1.6-alpha.1
2026-09-15 10:42:33 +08:00
creatixchu ea53423b60 release(dsh): 0.1.6-alpha.1 2026-09-15 09:35:57 +08:00
Chinesezjc 64a8ba929d Merge pull request #4195 from deepseek-harness/fix/compaction-banner-offset
fix(web): hold a compaction summary's code banner below the pinned header
2026-09-15 00:13:26 +08:00
Chinesezjc b416ac99e4 fix(web): hold a compaction summary's code banner below the pinned header
A summary's fenced code block pins its banner at the scrollport top, where the
pinned compaction header already sits, so the header covers the banner's Copy
control and a click there collapses the summary. Stop the banner below the
header's band and square the pinned header's corners so its opaque fill masks
the scrolling prose at the edges.
2026-09-15 00:03:33 +08:00
Chinesezjc c75e3d4e81 Merge pull request #1170 from deepseek-harness/sticky-collapsible-headers
feat(web): pin Think and compaction headers sticky while scrolling
2026-09-14 23:34:53 +08:00
imccyu 57364a8475 Merge pull request #4104 from deepseek-harness/feat/electron-asar-runtime-resolution
feat(desktop): use runtime profile resolution in packaged carriers
2026-09-14 23:30:37 +08:00
imccyu 344a3c9257 test(desktop): avoid unsafe asymmetric matchers 2026-09-14 23:13:01 +08:00
imccyu 01f2f01a2f Merge pull request #3651 from deepseek-harness/fix/profile-module-resolution
feat(boot): add profile resolution modes
2026-09-14 23:12:15 +08:00
imccyu 1b56c865fa fix(pkg): restore VFS Dirent reads 2026-09-14 23:10:47 +08:00
imccyu fa7d5519f5 feat(desktop): run runtime host from asar 2026-09-14 22:52:44 +08:00
imccyu c917a4e0bd feat(cli): force runtime resolution in pkg builds 2026-09-14 22:52:44 +08:00
imccyu 0bade5a010 chore(runtime): upgrade builtin loader to 0.1.6 2026-09-14 22:52:44 +08:00
imccyu c3a66d9cd2 refactor(boot): use addon-managed native cache 2026-09-14 22:50:11 +08:00
imccyu d4350442e8 docs(boot): remove private native cache lifecycle 2026-09-14 22:50:00 +08:00
imccyu 0de61878b3 fix(boot): treat null exports as legacy resolution 2026-09-14 22:30:51 +08:00
imccyu 32e6e872f6 fix(boot): bind internal ESM default resolver 2026-09-14 22:30:51 +08:00
imccyu 0d4b097375 fix(app-boot): align remaining runtime resolution edges 2026-09-14 22:30:51 +08:00
imccyu 69ab6575d1 fix(boot): preserve selected explicit path errors 2026-09-14 22:30:50 +08:00
imccyu 3cacc0e69a fix(boot): preserve selected package failures 2026-09-14 22:30:50 +08:00
imccyu c75f667822 fix(boot): cover profile resolution edge cases 2026-09-14 22:30:50 +08:00
imccyu db098c8a24 fix(presets): isolate package lookup failures 2026-09-14 22:30:50 +08:00
imccyu 013aead5c9 fix(boot): align runtime profile resolution with links 2026-09-14 22:30:50 +08:00
imccyu 2fc699255f test(boot): use native path canonicalization 2026-09-14 22:30:49 +08:00
imccyu a663bbf4b8 test(boot): canonicalize Windows resolver paths 2026-09-14 22:30:49 +08:00
imccyu 9c914dfea9 test(boot): cover explicit fallback failures 2026-09-14 22:30:49 +08:00
imccyu 6823f8a00c fix(boot): align runtime and link resolution 2026-09-14 22:30:49 +08:00
imccyu aa4c46edeb fix(boot): drop stale workflow worker dependency 2026-09-14 22:30:49 +08:00
imccyu 123272b05e chore(cli): restore profile resolver lockfile entry 2026-09-14 22:30:49 +08:00
imccyu 662bdf3833 docs(profile-resolution): clarify review assumptions 2026-09-14 22:30:48 +08:00
imccyu 6aa2e4633c feat(boot): add profile resolution modes 2026-09-14 22:30:48 +08:00
imccyu 22fed0f1eb Merge pull request #4191 from deepseek-harness/worktree-draft-editor-isolation
refactor(web): isolate draft editor runtime and view bindings
2026-09-14 22:24:45 +08:00
imccyu b5abe6cf29 refactor(web): isolate draft editor implementation without behavior changes 2026-09-14 22:10:52 +08:00
imccyu 218db70b75 docs(web): define two-stage draft editor isolation 2026-09-14 22:10:52 +08:00
_Kerman 11c4c52d1a Merge pull request #4186 from deepseek-harness/xtr/revert-3880-chat-behavior
fix(chat): revert Chat behavior changes from #3880
2026-09-14 22:03:27 +08:00
_Kerman d7a83ca5d6 docs(trajectory): clarify timing and thinking ownership 2026-09-14 21:55:14 +08:00
_Kerman 24a4c88859 test(chat): pin restored timing through settlement and reload 2026-09-14 21:55:05 +08:00
CreatixChu db942b3530 Merge pull request #4183 from deepseek-harness/worktree/session-log-upload-default-test-aware
feat(session): enable log upload by default with explicit test configuration
2026-09-14 21:49:42 +08:00
creatixchu 0eb87a4590 test: fix upload expectations and exited process cleanup 2026-09-14 21:34:38 +08:00
Tianyi Cui 7da882dbeb fix(session): pin the disabled upload branch in the real-API extension test
With the plugin defaulting to on, the messages e2e case that asserts an
absent `dsh_session_log` must disable it explicitly; otherwise the stub
rejects the field it did not expect and the model call finishes as an
error. Also align the wire reference and feedback subsystem docs with the
default-on field.
2026-09-14 21:27:27 +08:00
creatixchu 9532ed3b28 test(sdk): declare upload policy for existing replay compositions 2026-09-14 21:19:03 +08:00
creatixchu 31ec6bc7e3 fix(session): keep upload defaults independent of test runners 2026-09-14 21:15:42 +08:00
creatixchu ac917f4df3 docs: align Chinese session-log source reference 2026-09-14 21:04:13 +08:00
creatixchu 4885f90769 docs: refresh session-log config source reference 2026-09-14 21:02:31 +08:00
_Kerman 251ef01acf fix(chat): revert historical timing recovery from #3880 2026-09-14 21:00:40 +08:00
creatixchu 32b9c99560 Merge remote-tracking branch 'origin/worktree/session-log-upload-default-test-aware' into codex/pr-4183-ci 2026-09-14 21:00:20 +08:00
_Kerman 3db1085da2 fix(chat): restore collapsed thinking by default 2026-09-14 20:59:55 +08:00
creatixchu b35d728d42 test(session): cover environment-aware upload defaults 2026-09-14 20:59:55 +08:00
creatixchu ebf89a8455 Merge remote-tracking branch 'origin/master' into codex/pr-4183-ci 2026-09-14 20:59:46 +08:00
Tianyi Cui 6ce915d3fc fix(session): keep the recorded-session marker check branch-free
The per-file coverage gate measures `packages/*/*/src` at 100% branches;
a short-circuiting `||` over the two lane markers leaves the second
operand unexecuted under Vitest. Testing the marker list instead keeps
the same default without an uncovered branch.
2026-09-14 20:58:49 +08:00
Tianyi Cui 2389b65246 feat(session): default session-log upload on outside recorded-session lanes
`@deepseek-ai/dsh-session-log-deepseek` now contributes `dsh_session_log`
unless a composition sets `enabled: false`. A recorded-Session lane
(`$VITEST` or `$DSH_SNAPSHOT`) keeps the default off, because its
committed fixtures are both replay input and expected output for the
persisted log; no fixture regeneration is needed.
2026-09-14 20:49:44 +08:00
Yichen Jiang 4f7336c831 Merge pull request #4178 from deepseek-harness/worktree/terminal-color-contrast
fix(web): preserve terminal colors and cursor contrast
2026-09-14 20:38:25 +08:00
Yichen Jiang 578e1e428f test(web): cover terminal cursor theme binding 2026-09-14 20:30:19 +08:00
Yichen Jiang 80043785c4 fix(web): preserve terminal colors and cursor contrast 2026-09-14 20:17:19 +08:00
Yichen Jiang 3cbc19d1d0 Merge pull request #4173 from deepseek-harness/worktree/sidebar-guide-terminal-menu
feat(web): launch terminals from provider guide menus
2026-09-14 19:52:22 +08:00
Chinesezjc 67271a921b feat(web): pin Think and compaction headers sticky while scrolling
Both blocks render their expanded body uncapped, so a long reasoning chain or
compaction summary carried the disclosure header off the viewport and buried
the toggle. Pin each header to the conversation scroll container while the
block is open, keeping the rule scoped to these two blocks so the capped tool
rows do not stack headers at the top of the viewport.

The compaction header also overrides its stacking rank above the summary's
code-block banner and keeps an opaque hover fill.
2026-09-14 19:52:04 +08:00
Yichen Jiang 4012092d24 feat(web): launch terminals from provider guide menus 2026-09-14 19:22:22 +08:00
Turtle ad9fd17faf Merge pull request #4164 from deepseek-harness/turtle/orchestrator-3d7fcb93e79d
fix(client): show local changes in edit diff cards
2026-09-14 19:11:08 +08:00
Turtle f50309b5d9 fix(client): bound contextual diff comparison work 2026-09-14 19:02:59 +08:00
Yifffan 7c4067c24e Merge pull request #4044 from deepseek-harness/fix/sidebar-document-preview-polish
Sidebar refine: 侧边栏 PDF/图片预览/格式空态/loading 等整体优化
2026-09-14 18:37:30 +08:00
Turtle 53786d6415 test(client): guard contextual diff replay against page errors 2026-09-14 18:23:58 +08:00
Yif f51addaa0d fix(web-sidebar): resolve review feedback on document preview polish
- files tree records its scroll offset in a ref and commits it once on
  unmount, only while the owner's signal is live, so a scroll neither
  re-renders the tree nor writes to a forgotten bucket
- LoadingIndicator drops iconOnly and always renders icon-only
- registry and the unviewable list share one suffix-matching module, and
  register() rejects binaryExtensions entries outside extensions
- drop the ambiguous key suffix from the unviewable office group
- extract the text preview's header path block shared by the unsupported
  and content branches; clsx for static class joins; CSS trailing newline
- sync the preview-operations and polish Agent Notes with shipped facts
2026-09-14 18:06:23 +08:00
Ruilin Geng bb2b325486 Merge pull request #4137 from deepseek-harness/feat/docs-site-mermaid-viewer
feat(docs): 支持 Mermaid 图表全屏缩放和平移
2026-09-14 17:57:00 +08:00
Turtle 242410cf4b fix(client): render actual changes in contextual diff cards 2026-09-14 17:52:28 +08:00
Turtle e32f88dff2 Merge pull request #4024 from deepseek-harness/turtle/orchestrator-837dec7391c5
perf(terminal-bash): make scrollback retention incremental
2026-09-14 17:05:02 +08:00
Yichen Jiang f14032abd3 Merge pull request #4155 from deepseek-harness/worktree/web-terminal-theme
feat(web): follow application theme in sidebar terminals
2026-09-14 17:01:40 +08:00
Turtle b42913c8f1 Merge master and adapt terminal retention fixtures 2026-09-14 16:52:55 +08:00
Yifffan 44cee2e1f5 Merge branch 'master' into fix/sidebar-document-preview-polish 2026-09-14 16:35:23 +08:00
Tianyi Cui 0e9e89e63f Merge pull request #4149 from deepseek-harness/worktree/projection-cache-drain-20260914
test: wait for cache writes and workflow child starts
2026-09-14 16:35:01 +08:00
Tianyi Cui b730c2cb83 Merge pull request #4092 from deepseek-harness/worktree/browser-use-backends
feat(browser-use): add experimental per-Session backends
2026-09-14 16:33:14 +08:00
Yichen Jiang 1daed09185 feat(web): follow application theme in sidebar terminals 2026-09-14 16:18:50 +08:00
Tianyi Cui 7e8d42d122 test(workflow): await host child starts before cancellation 2026-09-14 16:17:59 +08:00
Tianyi Cui 88e0a5acd1 test(session): await every successful cache replacement 2026-09-14 16:17:41 +08:00
Tianyi Cui 3ba25cd4cb test(session): exclude interval fallback from mandatory checkpoint 2026-09-14 16:09:15 +08:00
Tianyi Cui a90d8df582 test(session): await durable turn-end cache checkpoints 2026-09-14 16:00:09 +08:00
Tianyi Cui a3fb17c703 Merge pull request #4141 from deepseek-harness/worktree/subagent-settlement-text
fix(subagent): keep settlement notices text-only
2026-09-14 15:50:22 +08:00
Tianyi Cui 16f97e2dfb refactor(browser-use): defer DSH inference integration 2026-09-14 15:43:14 +08:00
Yichen Jiang 0ef4ff9bd0 Merge pull request #4139 from deepseek-harness/worktree/terminal-common-shells
feat(web): show common shells by default
2026-09-14 15:36:15 +08:00
Tianyi Cui b86b89da94 docs(subagent): align settlement notice descriptions 2026-09-14 15:27:38 +08:00
Tianyi Cui 1d26330415 test(subagent): reuse Messages fixtures and clarify notice contracts 2026-09-14 15:13:13 +08:00
Tianyi Cui 6caeb505c5 fix(browser-use): await existing Agent creation lifecycle 2026-09-14 14:50:23 +08:00
gengruilin 91f6733096 fix(docs): align Mermaid viewer accessible titles 2026-09-14 14:49:48 +08:00
Tianyi Cui e543cc9a6e Merge awaited Agent creation from master 2026-09-14 14:46:15 +08:00
Tianyi Cui 29debb8b24 fix(subagent): keep settlement notices text-only 2026-09-14 14:44:11 +08:00
Tianyi Cui ebe3ee4d4c docs(browser-use): refresh auxiliary event type acknowledgement 2026-09-14 14:33:23 +08:00
Tianyi Cui 52a8f7a759 docs(browser-use): clarify conditional tool ordering 2026-09-14 14:31:53 +08:00
Yichen Jiang 1ddd7e08c3 fix(web): keep Windows PowerShell in terminal choices 2026-09-14 14:31:41 +08:00
Tianyi Cui ee031f9cf9 Merge master into browser-use backends 2026-09-14 14:30:58 +08:00
_Kerman 28a4241e9f Merge pull request #3583 from deepseek-harness/xtr/agent-setup-hook
feat(agent)!: replace agent/session-start with awaited agent/created
2026-09-14 14:25:09 +08:00
Tianyi Cui 9845c543c0 docs(browser-use): refresh provider dependency graph 2026-09-14 14:23:52 +08:00
Yichen Jiang 249c8d1565 feat(web): show common shells by default 2026-09-14 14:20:25 +08:00
07akioni 49fcf87cb2 Merge pull request #4133 from deepseek-harness/07akioni/fix-table-hover-height
fix(web): prevent short markdown tables from shifting on hover
2026-09-14 14:03:49 +08:00
_Kerman d3856902b6 Merge pull request #3880 from deepseek-harness/xtr/trajectory-json-string-wrapping
feat(client): improve trajectory JSON and PTC inspection
2026-09-14 14:01:59 +08:00
_Kerman 0a382becb3 Merge latest origin/master into xtr/agent-setup-hook 2026-09-14 13:57:18 +08:00
07akioni f0463cf80d fix(test): use block bodies for void browser callbacks 2026-09-14 13:53:09 +08:00
gengruilin ece25ff8e6 refactor(docs): simplify Mermaid fullscreen viewer controls 2026-09-14 13:52:42 +08:00
_Kerman 1119fa2544 test(web): refresh the Auto-review trajectory golden for PTC summaries
The trajectory ledger now summarizes a recorded run_code call with the
program description instead of its raw JSON arguments, and marks it with
the code-inspection icon. Refresh the Auto-review denial golden so the
replayed ARIA expectation matches the shipped rendering.
2026-09-14 13:50:23 +08:00
Yichen Jiang 444c801718 Merge pull request #3864 from deepseek-harness/worktree/web-sidebar-terminal
feat(web): add interactive sidebar terminals
2026-09-14 13:47:12 +08:00
07akioni 8ab3579e06 test(web): guard fitting table geometry across interaction 2026-09-14 13:34:00 +08:00
Turtle 6a4a825a88 Merge pull request #4079 from deepseek-harness/turtle/master-ci-platform-tests
test(ci): repair platform fixtures, gallery clock and cache teardown
2026-09-14 12:39:59 +08:00
gengruilin cd0cc91382 fix(docs): keep Mermaid wheel zoom anchored to the pointer 2026-09-14 12:36:32 +08:00
gengruilin b0ec8cef32 feat(docs): add fullscreen Mermaid diagram viewer 2026-09-14 12:25:23 +08:00
Yichen Jiang 0a0563cc9b test(web): mock retained terminals in assembled fixtures 2026-09-14 12:13:26 +08:00
Yichen Jiang 0a7f47e3f2 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal 2026-09-14 12:03:02 +08:00
Yichen Jiang 11b2114786 fix(web): align terminal launcher controls and guide appearance 2026-09-14 12:00:29 +08:00
07akioni 6d81d0fdbd test(web): update table hover scrollbar expectation 2026-09-14 11:47:13 +08:00
07akioni 55a17d2e57 fix(ui): keep markdown table height stable on hover 2026-09-14 11:46:02 +08:00
CreatixChu 48de2d1ba0 Merge pull request #3440 from deepseek-harness/worktree/3041-image-offload-log-design
feat(session, llm): durable image offload watermark
2026-09-14 11:17:06 +08:00
_Kerman a85778448a Merge origin/master into xtr/trajectory-json-string-wrapping 2026-09-14 11:15:34 +08:00
Yichen Jiang 3219a2d8e1 docs(api): refresh terminal dependency graph 2026-09-14 11:10:18 +08:00
Yichen Jiang 37e5d8c4aa fix(web): remember shell choices before terminal startup 2026-09-14 11:02:53 +08:00
Yichen Jiang 33d89aee77 feat(web): choose and remember terminal shells 2026-09-14 10:59:46 +08:00
Yichen Jiang 1b7b50bbe1 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal
# Conflicts:
#	apps/web/package.json
#	docs/config-catalog.i18n.yaml
#	docs/config-catalog.md
#	docs/event-producer-consumer.i18n.yaml
#	docs/event-producer-consumer.md
#	docs/event-producer-consumer.zh.md
#	packages/api/remotes/src/client/index.ts
#	packages/e2b/subprocess-e2b/README.md
#	packages/e2b/subprocess-e2b/README.zh.md
#	packages/e2b/subprocess-e2b/src/index.ts
#	packages/e2b/subprocess-e2b/src/terminal.ts
#	packages/e2b/subprocess-e2b/tests/subprocess.spec.ts
#	packages/e2b/subprocess-e2b/tests/terminal.spec.ts
#	packages/mcp/mcp-resources/README.i18n.yaml
#	pnpm-lock.yaml
#	scripts/verify-package-readme-model-experience.ts
#	snapshots/session/cordis-inspect-jsdoc/session.v3.jsonl
#	tsconfig.base.json
#	tsconfig.client.json
#	tsconfig.host.json
2026-09-14 10:36:31 +08:00
creatixchu 2b1dabec86 Refresh merged SDK recordings and persistence documentation 2026-09-14 10:25:54 +08:00
creatixchu 27d52a4edb Update image tests for split adapters and durable offload failures 2026-09-14 10:21:54 +08:00
creatixchu a7070008e7 Merge master and integrate image offload with both DeepSeek protocols 2026-09-14 10:19:11 +08:00
imccyu 9ccd9b0634 Merge pull request #3960 from deepseek-harness/worktree-feqatest
test(client): replace legacy fixtures with RemoteMock
2026-09-13 23:09:09 +08:00
imccyu f258024b4d docs(testing): correct fixture migration guidance 2026-09-13 22:51:53 +08:00
imccyu e179c4ae14 test(client): preserve migrated fixture semantics 2026-09-13 22:51:53 +08:00
imccyu c0e8252afe refactor(client): remove fixture-only API residue 2026-09-13 22:51:53 +08:00
Tianyi Cui b84a8226bd fix(browser-use): initialize MCP clients without new core APIs 2026-09-13 22:39:18 +08:00
Tianyi Cui e41511d844 fix(browser-use): integrate current MCP resources and snapshots 2026-09-13 22:39:17 +08:00
Tianyi Cui d1e32e481a fix(browser-use): close lifecycle gaps and isolate replay fixtures 2026-09-13 22:39:16 +08:00
Tianyi Cui 0752a09ae6 docs(browser-use): explain the Stagehand action timer limit 2026-09-13 22:39:16 +08:00
Tianyi Cui e1612c2fdc feat(browser-use): add per-Session experimental browser backends 2026-09-13 22:39:16 +08:00
imccyu 7e331d72ed test(file-upload): cover unnamed byte uploads 2026-09-13 21:54:31 +08:00
imccyu 566db0b6cb docs(testing): retire browser fixture references 2026-09-13 21:30:03 +08:00
imccyu 5541ae0d2d refactor(client): replace browser fixture with RemoteMock 2026-09-13 21:30:03 +08:00
imccyu eb2cb668a5 test(web): run fixture browser cases through Host 2026-09-13 21:30:02 +08:00
imccyu 9b091919e2 test(session-controller): migrate remaining Remote fixtures 2026-09-13 21:30:02 +08:00
imccyu c1504a2aa6 fix(client-modules): resolve the owning Loader instance 2026-09-13 21:27:54 +08:00
imccyu 77db56687b test(ui-message-feedback): use RemoteMock 2026-09-13 21:27:53 +08:00
imccyu a475a65bed fix(client-test-runtime): isolate client instances 2026-09-13 21:27:53 +08:00
Tianyi Cui 5ca77db865 Merge pull request #4089 from deepseek-harness/worktree/deepseek-messages-parity
feat(llm): default DeepSeek to Messages with Files parity
2026-09-13 21:19:03 +08:00
Tianyi Cui accafa5fb2 refactor(llm): inherit Messages defaults and verify live parity 2026-09-13 20:36:50 +08:00
Tianyi Cui aef56a3108 test(ci): align external-agent and workflow endpoint fixtures 2026-09-13 20:36:50 +08:00
Tianyi Cui ece35a1a38 test(ci): exercise Messages defaults in runtime smokes 2026-09-13 20:36:50 +08:00
Tianyi Cui 7d3dab66a2 fix(llm): harden Messages transport and Files parity 2026-09-13 20:36:50 +08:00
Tianyi Cui b0641b83fc feat(llm): default DeepSeek to Messages with Files parity 2026-09-13 20:36:50 +08:00
Tianyi Cui 9531265ff3 Merge pull request #4098 from deepseek-harness/worktree/unarchive-sessions
feat(workspace): restore archived sessions from a settings page
2026-09-13 17:36:52 +08:00
Tianyi Cui 76941d0085 fix(workspace): keep the newest archive set from stale replies
Both archive verbs answer with the complete set, so a reply that lands
after a newer request or after a pushed increment overwrote it. The
Client model now installs an answer only while it is still the latest
archive-set request, and the follow increment and baseline supersede it.

Also narrow the page's public surface and its states: the locale
namespace stays internal, row derivation is a memo over the framework
hooks, the search input carries one accessible name, the empty-archive
message no longer covers a set whose entries cannot be restored, and the
new package declares no client dependency edge it does not import.
2026-09-13 16:45:45 +08:00
Tianyi Cui e1ee806226 chore(cordis): regenerate the client slot catalog
The new Settings section registers one more settings.section contribution,
which the generated client slot catalog lists.
2026-09-13 16:45:45 +08:00
Tianyi Cui ae34320a5b test(web): expect the unarchive verb and the archived-sessions nav row
The Client Cordis catalog now publishes unarchiveSession on both the
Workspace service and the UI navigation capability, and the new Settings
section adds one nav row that every settings-dialog aria snapshot carries.
Refresh those expectations; the goldens gain only that row.
2026-09-13 16:45:45 +08:00
Tianyi Cui 5f773a0ded feat(workspace): restore archived sessions from a settings page
Sessions could be archived from the sidebar row menu but never restored:
the registry-global archive set was a one-way display filter with no
inverse RPC and no surface listing its members.

WorkspaceRegistry.unarchiveSession drops one id from archivedSessionIds in
one durable write on the same operation chain as archive, with no
session-existence probe because a removal cannot introduce an unknown
referent. The Workspace controller exposes it as workspace.unarchiveSession
returning the complete archive set, which the existing archived follow
increment already propagates, so no frame type and no persisted field
changes.

The new client package dsh-client-ui-settings-unarchive-sessions owns the
restore surface: one localized settings.section page listing the archive
set joined with loaded session summaries, newest archive first, searchable
by title or workspace, with one Unarchive action per row.
2026-09-13 16:45:45 +08:00
Tianyi Cui 9e87d16f3a Merge pull request #4116 from deepseek-harness/worktree/coverage-attribution
fix(coverage): canonicalize partition locations before the blob merge
2026-09-13 16:45:02 +08:00
Tianyi Cui 8ba386ec69 Merge branch 'master' into worktree/coverage-attribution 2026-09-13 16:37:55 +08:00
Tianyi Cui 2214891606 Merge pull request #4119 from deepseek-harness/worktree/workflow-sandbox-ptc
fix(workflow): execute orchestration through sandboxed PTC
2026-09-13 16:33:51 +08:00
Tianyi Cui 10930a0205 fix(coverage): fail loud when the canonicalizer gets no coverage data
Vitest types `Reporter.onCoverage` as `unknown`, so a payload that carries no
istanbul `data` record now fails the partition instead of silently leaving
every location uncanonicalized; a spec case pins that rejection. The module
docs keep the contract and link the Agent Note for the mechanism, the
partition spec names what it asserts (the reporter argument, not Vitest hook
ordering), and the notes correct the `Infinity` origin to ast-v8-to-istanbul,
record the reporter-ordering dependency, and name the reporter each partition
child carries.
2026-09-13 15:17:06 +08:00
Tianyi Cui c171a7a9a2 fix(coverage): canonicalize partition locations before the blob merge
A source file carries one istanbul statement map per Vite environment, so a
node suite and a jsdom suite that load the same module enter the merged
report through two maps that spell one statement differently. istanbul
reconciles those overlapping ranges by attributing the hits of the narrowest
containing range, but only while a location carries numeric line and column
values. ast-v8-to-istanbul ends a whole-line statement at column Infinity, and
the partition blob serializes Infinity as null, so the merge command kept the
client-only spelling as an extra, unhit statement: a file whose every statement
ran reported 96.96% and failed the per-file 100% gate.

The coordinator now passes each partition a reporter whose onCoverage hook
rewrites every non-finite end column to a finite line-end column before the
blob is written, which restores that reconciliation without weakening the
gate. The uncovered-locations reporter reads the same column as a line end.
2026-09-13 15:13:59 +08:00
Tianyi Cui 9b6457bc12 Merge pull request #4106 from deepseek-harness/worktree/mcp/optional-surface
feat(mcp): expose profile resources only when servers are configured
2026-09-13 14:59:10 +08:00
Tianyi Cui e9a300f192 Merge pull request #4100 from deepseek-harness/worktree/persistence-format-history
docs(session): backfill and require every persistence format reference
2026-09-13 14:58:05 +08:00
Tianyi Cui e636da11fb Merge branch 'master' into worktree/mcp/optional-surface 2026-09-13 14:52:01 +08:00
Tianyi Cui 6811c59efb Merge branch 'master' into worktree/persistence-format-history 2026-09-13 14:49:15 +08:00
Tianyi Cui 34f80b8e6e fix(workflow): close lifecycle and CI integration gaps 2026-09-13 14:47:56 +08:00
Tianyi Cui 35af8698c2 fix(workflow): execute orchestration in the sandboxed PTC runtime 2026-09-13 14:47:56 +08:00
Tianyi Cui 4f071d6762 Merge pull request #4117 from deepseek-harness/worktree/fix-mcp-snapshot-ralph-defaults
test(mcp): align snapshots with Ralph disabled by default
2026-09-13 14:47:53 +08:00
Yif f7e010a70a docs(agent-notes): cite issue 3974 as plain text to satisfy the repository-references gate 2026-09-13 13:37:02 +08:00
Yif 25e6e976b7 test(sidebar-documentpreview): cover the visible-label loading branch and drop the stale single-viewer control from the seeded-history golden 2026-09-13 13:28:16 +08:00
Yifffan 20d78f0e77 Merge branch 'master' into fix/sidebar-document-preview-polish 2026-09-13 07:26:09 +02:00
Tianyi Cui 004ef35010 test(mcp): align snapshots with Ralph disabled by default 2026-09-13 12:39:33 +08:00
Tianyi Cui 38070996c1 docs(session): nest historical formats under persistence changes 2026-09-13 12:19:46 +08:00
Tianyi Cui 765c87fb87 fix(docs): address persistence format history review 2026-09-13 12:08:08 +08:00
Tianyi Cui b08f6c421b docs(session): backfill and gate persistence format references 2026-09-13 12:08:07 +08:00
Tianyi Cui b5376e3d02 test(mcp): align resource snapshots with shipped defaults 2026-09-13 11:55:25 +08:00
Tianyi Cui 4c6b2c02af docs(mcp): clarify resource discovery and profile verification 2026-09-13 11:52:42 +08:00
Tianyi Cui 815a5eb01e test(mcp): include resources in minimal profile expectations 2026-09-13 11:52:42 +08:00
Tianyi Cui e08468954a feat(mcp): activate profile resource tools for configured servers 2026-09-13 11:52:42 +08:00
Tianyi Cui 99a73ca1d4 Merge pull request #4101 from deepseek-harness/worktree/demote-ralph
feat(presets): disable ralph in the default compositions
2026-09-13 11:42:40 +08:00
Tianyi Cui 351645a6c9 Merge pull request #4112 from deepseek-harness/worktree/file-search-root-fixture-20260913
test(file-reference): publish restored workspace fixtures atomically
2026-09-13 11:34:01 +08:00
Tianyi Cui a9bba3bac5 Merge pull request #4082 from deepseek-harness/worktree/mcp/resources
feat(mcp): add scoped resources and server instructions
2026-09-13 11:27:47 +08:00
Tianyi Cui deb0d7e1b5 test(file-reference): publish restored workspace fixtures atomically 2026-09-13 04:56:35 +08:00
Tianyi Cui c7d2e99b8e docs(presets): state the web-app disable in the present tense 2026-09-13 01:14:06 +08:00
turtle1999 c8370240a8 test(ci): pin gallery time and dispose projection cache fixtures 2026-09-13 00:42:12 +08:00
Tianyi Cui d49192b6ac docs(presets): name the new id in the ralph restore recipe 2026-09-13 00:41:44 +08:00
turtle1999 d9ec28ed3b Merge branch 'master' of https://github.com/deepseek-harness/deepseek-harness into turtle/master-ci-platform-tests 2026-09-13 00:37:52 +08:00
Tianyi Cui 73985344cd feat(presets): disable ralph in the default compositions 2026-09-13 00:15:39 +08:00
Tianyi Cui 7efd4ab20c Merge pull request #4087 from deepseek-harness/worktree/ci/dependency-catalog-rescope
fix(ci): preserve recorded npm resolution during vendor rescoping
2026-09-12 23:25:44 +08:00
Tianyi Cui 834bd55a59 fix(mcp): expose scoped resource server names to the model 2026-09-12 22:40:51 +08:00
Tianyi Cui aec2d5e5f0 docs(mcp): explain resource providers and server instructions 2026-09-12 22:40:51 +08:00
Tianyi Cui 3ba5b6eb04 feat(mcp): add scoped resources and server instructions 2026-09-12 22:40:50 +08:00
Tianyi Cui f6e7166762 docs(mcp): align bridge note with SDK-owned discovery 2026-09-12 22:38:00 +08:00
Tianyi Cui 3785b68ecf fix(mcp): own transports through protocol negotiation 2026-09-12 22:31:42 +08:00
Tianyi Cui 395f79b772 docs(mcp): describe subsystem roles and configuration 2026-09-12 22:31:42 +08:00
Tianyi Cui 489c3ac715 feat(mcp): negotiate modern protocols with the official SDK
Preserve the verified SDK layer tree from c8b248b480 before rebasing its complete net change.
2026-09-12 22:31:42 +08:00
Tianyi Cui 55fc8b91e4 fix(ci): preserve recorded npm resolution during vendor rescoping 2026-09-12 22:25:10 +08:00
Tianyi Cui c2a68638b2 Merge pull request #4059 from deepseek-harness/worktree/publish-all-experimental
feat(release): publish all experimental packages
2026-09-12 22:19:29 +08:00
Tianyi Cui 60a413034a Merge branch 'worktree/default-product-no-experimental' into worktree/publish-all-experimental 2026-09-12 22:04:27 +08:00
Tianyi Cui a280cf34f5 test: canonicalize browser notices fixture paths 2026-09-12 22:04:16 +08:00
Tianyi Cui 7a4dca21c7 Merge branch 'worktree/default-product-no-experimental' into worktree/publish-all-experimental 2026-09-12 21:32:08 +08:00
Tianyi Cui 0e060aa675 fix(ci): canonicalize browser notice roots with native realpath 2026-09-12 21:30:51 +08:00
Tianyi Cui 00b2470eb6 test: canonicalize Windows Vite fixture roots 2026-09-12 21:26:53 +08:00
Tianyi Cui 5b491e8cc6 fix(ci): skip product graph capture during dependency analysis 2026-09-12 21:20:43 +08:00
Tianyi Cui b5d463ca5c test(release): cover packed experimental worker consumers 2026-09-12 20:56:26 +08:00
Tianyi Cui 22c5ae6abc fix: cover supported experimental isolation inputs 2026-09-12 20:56:26 +08:00
Tianyi Cui 42c5b65643 feat(release): publish all experimental packages 2026-09-12 20:56:26 +08:00
Tianyi Cui 8f14f98a1a ci(web): verify composed profiles, live registries, and bundle inputs 2026-09-12 20:56:26 +08:00
Tianyi Cui f34dd86285 ci(release): keep experimental packages out of the default product 2026-09-12 20:56:26 +08:00
Tianyi Cui 251210d517 test(release): cover baseline package discovery 2026-09-12 20:56:26 +08:00
Tianyi Cui 6e30e1a300 feat(release): use a private denylist for experimental packages 2026-09-12 20:56:26 +08:00
Tianyi Cui 5bb449f716 Merge pull request #3717 from deepseek-harness/worktree/pr2437-recovery/metadata
fix: include package-index links and align residual metadata
2026-09-12 20:52:51 +08:00
Tianyi Cui 3b27563290 fix: include package-index links and align residual metadata 2026-09-12 20:20:02 +08:00
Tianyi Cui 6c3c3064c3 refactor: align instruction and job helper symbols 2026-09-12 20:20:02 +08:00
Tianyi Cui 739285c813 docs: correct surviving package and service references 2026-09-12 20:20:02 +08:00
Tianyi Cui be682f7a59 Merge pull request #4086 from deepseek-harness/worktree/linux-signal-outcome-preservation
fix(subprocess): preserve exit outcomes when signaling fails
2026-09-12 20:12:20 +08:00
Tianyi Cui dd55f38614 docs(subprocess): record signal error isolation and fixture safety 2026-09-12 19:44:51 +08:00
Tianyi Cui 083d192007 Merge current master into direct outcome fix 2026-09-12 19:25:51 +08:00
Tianyi Cui 020b84c301 Merge pull request #4060 from deepseek-harness/worktree/persistence-prerelease-backfill
docs(session): backfill persistence history for all alpha and RC tags
2026-09-12 19:16:42 +08:00
Tianyi Cui 71334b9c78 fix(subprocess): preserve direct outcomes on signal failure 2026-09-12 19:14:34 +08:00
pku-xht 771dd49a06 Merge pull request #3304 from deepseek-harness/auto-review-per-call
feat(permission): add Auto review mode
2026-09-12 18:58:41 +08:00
Tianyi Cui 3b264679ea Merge pull request #4085 from deepseek-harness/worktree/linux-direct-signal-ack
fix(subprocess): confirm direct SIGKILL after group delivery
2026-09-12 18:57:14 +08:00
Tianyi Cui 4adfe00207 fix(session): validate generated release facts 2026-09-12 18:52:17 +08:00
Tianyi Cui 0b7f0b072b chore: enforce maintained repository reference policy 2026-09-12 18:52:17 +08:00
Tianyi Cui 2d5391eede docs(session): use tags throughout historical release records 2026-09-12 18:52:17 +08:00
Tianyi Cui aea077ac75 docs(session): backfill alpha and RC persistence history 2026-09-12 18:52:17 +08:00
Tianyi Cui 35dbd84ac4 test: isolate Git config without a Windows device path 2026-09-12 18:52:02 +08:00
Tianyi Cui c07df5aa65 fix(subprocess): confirm direct SIGKILL after group delivery 2026-09-12 18:49:15 +08:00
Tianyi Cui 2b367a9aeb Merge updated Computer Use parent with master CI fix 2026-09-12 18:40:48 +08:00
Tianyi Cui 7a841fa908 Merge master with subprocess CI fix from #4080 2026-09-12 18:39:39 +08:00
Tianyi Cui b905d399df Fix offline reference checks and preserve native pack inputs 2026-09-12 18:37:56 +08:00
Tianyi Cui 1b4e1ac076 test: separate independent policy gate registration cases 2026-09-12 18:37:56 +08:00
Tianyi Cui 6b651380a7 feat: enforce maintained repository reference policy 2026-09-12 18:37:56 +08:00
Tianyi Cui 631c52fd4f Merge master into computer-use provider branch 2026-09-12 18:37:00 +08:00
Tianyi Cui 76a4153330 Merge pull request #4080 from deepseek-harness/worktree/acp-scope-settlement-fix
fix(subprocess): wait for fallback exit before scope failure
2026-09-12 18:33:55 +08:00
Tianyi Cui a43fd7d8a7 Merge pull request #4071 from deepseek-harness/worktree/dependency-catalog
docs: generate deduplicated npm dependency JSON
2026-09-12 18:31:14 +08:00
Magolor a630265a02 docs(session): define the LogicalSession refactor plan (#3654)
* docs(session): define the runtime and storage refactor plan

Document the staged SessionRuntime, SessionService, and SessionStorage design, developer transition rules, capability protocols, data compatibility, and FAQ in English and Chinese.

* docs(session): describe the switchable runtime-access gate in the plan

The plan said the access gate rejects concrete-class bindings. The
shipped gate is switchable: its checked-in default is `warn`, which
prints every finding, annotates it in GitHub Actions, and exits 0, so
branches that still bind `Session` merge while their consumers migrate.
`DSH_SESSION_RUNTIME_ACCESS=enforce` fails the gate for one run, and the
checked-in default flips to `enforce` once no in-flight branch binds
the class. Record that in the stage table and the enforcement section.

* docs(session): rename SessionRuntime to LogicalSession in the rebuild plan

The logical session interface is named `LogicalSession`; the plan, the
capability protocols, the developer transition rules, and the FAQ use
that name for the type, the detached factories (`createLogicalSession`,
`restoreLogicalSession`), the access gate (`verify-logical-session-access`,
`DSH_LOGICAL_SESSION_ACCESS`), and the prose that called the object a
runtime. The overview note moves to
`2026-09-06-logical-session-storage-rebuild.*` with a new pairing record.

The FAQ records why `SessionRuntime` was rejected: the package naming
table reserves `Runtime` for a component that runs live work, and the
interface holds one session's identity and log without running anything.
2026-09-12 18:11:15 +08:00
Tianyi Cui c6721f992e docs: retain hosted run identities without organization URLs 2026-09-12 18:05:41 +08:00
Tianyi Cui 4b99ace8fb fix(subprocess): await direct fallback settlement before scope failure 2026-09-12 17:55:53 +08:00
turtle1999 fabbcf8c76 test(ci): cover control requests and fix cross-volume fixtures 2026-09-12 17:38:12 +08:00
Tianyi Cui 66734e344a Merge Computer Use parent into Auto review for PR stack 2026-09-12 17:06:46 +08:00
Tianyi Cui 7900404541 fix: deduplicate dependency JSON and isolate npm configuration 2026-09-12 17:06:38 +08:00
Tianyi Cui cd97419e81 fix(computer-use): cancel image admission and fix source snapshots 2026-09-12 16:43:52 +08:00
Tianyi Cui af4ad05845 feat: add computer use with Cua Driver providers 2026-09-12 16:41:12 +08:00
Tianyi Cui 313768f520 docs: generate published npm dependency catalog 2026-09-12 16:36:36 +08:00
Tianyi Cui 03888a1619 Merge pull request #4066 from deepseek-harness/turtle/master-ci-persistence-budget
test(ci): stabilize master persistence integration budgets
2026-09-12 16:09:09 +08:00
Tianyi Cui a92ac9e949 Merge pull request #4057 from deepseek-harness/worktree/ptc-sandbox/ptc-naming
refactor(ptc): align runtime packages and services with PTC naming
2026-09-12 16:05:11 +08:00
Tianyi Cui 83d121c8c0 docs(ptc): align runtime terminology and naming decision links 2026-09-12 15:03:04 +08:00
Tianyi Cui 7c9bb5914c refactor(ptc): align runtime packages and services with PTC naming 2026-09-12 15:03:04 +08:00
Tianyi Cui e892c1abf5 test(ssh): preserve stream reads across supported Node versions 2026-09-12 15:03:04 +08:00
Tianyi Cui 305847127a fix(deps): retain locked PDF canvas optional dependencies 2026-09-12 15:03:03 +08:00
Tianyi Cui 28c4daa66c fix(ssh): keep completion cleanup under release ownership 2026-09-12 15:03:03 +08:00
Tianyi Cui ac43c2656f fix(ssh): propagate output closure and bound exit observations 2026-09-12 15:03:03 +08:00
Tianyi Cui d3fa7030ef fix(ssh): retain failed results and join all process cleanup 2026-09-12 15:03:03 +08:00
Tianyi Cui d951a49038 fix(ssh): preserve remote filenames in file URLs 2026-09-12 15:03:03 +08:00
Tianyi Cui 119d44fc4c docs(ssh): generate the shared-instance dependency graph 2026-09-12 15:03:03 +08:00
Tianyi Cui c886f26b91 docs(ssh): record the administrative timer range 2026-09-12 15:03:03 +08:00
Tianyi Cui 1c5a201af8 fix(ssh): reject timeouts above the native timer range 2026-09-12 15:03:03 +08:00
Tianyi Cui 7673f90d18 docs(ssh): define trusted runtime installation prerequisites 2026-09-12 15:03:03 +08:00
Tianyi Cui 969e968e0e fix(ssh): disable signal-triggered helper inspection 2026-09-12 15:03:03 +08:00
Tianyi Cui c6e531e45b refactor(ssh): retain TLS ownership of raw socket errors 2026-09-12 15:03:03 +08:00
Tianyi Cui b5fd914306 test(ssh): isolate backend availability from helper dispatch 2026-09-12 15:03:03 +08:00
Tianyi Cui 561bfc3bbb fix(ssh): join authenticated stream teardown and validate package closure 2026-09-12 15:03:03 +08:00
Tianyi Cui abb7c6a49d fix(ssh): join transport cleanup and preserve collected output 2026-09-12 15:03:03 +08:00
Tianyi Cui 32de4412dc refactor(sandbox): await cancellable process preparation 2026-09-12 15:03:03 +08:00
Tianyi Cui 72226bd061 feat(ssh): authenticate streams and join remote process cleanup 2026-09-12 15:03:03 +08:00
Tianyi Cui 4fb0fdac68 feat(ssh): checkpoint POSIX helper and remote providers 2026-09-12 15:03:03 +08:00
Tianyi Cui 73e38e1758 fix(shell): include confinement preparation in foreground deadlines 2026-09-12 15:03:03 +08:00
Tianyi Cui f72c2e019d fix(preview): provide the native realpath callback alias 2026-09-12 15:03:03 +08:00
Tianyi Cui fd9d333fbe test(runtime): cover async preparation in built and Windows paths 2026-09-12 15:03:03 +08:00
Tianyi Cui 58b67dcb5a fix(fs): preserve physical traversal in directory entry paths 2026-09-12 15:03:03 +08:00
Tianyi Cui 0a7a69ec54 test(fs): exercise byte reads with a live cancellation signal 2026-09-12 15:03:02 +08:00
Tianyi Cui 4c7b6902e3 fix(fs): share native path rooting between resolve and lstat 2026-09-12 15:03:02 +08:00
Tianyi Cui 573f525c33 fix(fs): preserve native Windows drive-relative path resolution 2026-09-12 15:03:02 +08:00
Tianyi Cui caa69608fb refactor(sandbox): await cancellable preparation in process consumers 2026-09-12 15:03:02 +08:00
Tianyi Cui fc7f7ddda6 test(ptc): refresh portable denial and shared runtime schema fixtures 2026-09-12 15:03:02 +08:00
Tianyi Cui 4d4e333676 test(ptc): make sandbox denial recording portable and cover execution descriptors 2026-09-12 15:03:02 +08:00
Tianyi Cui a29991144a test(ptc): include program approval scenario in host type checking 2026-09-12 15:03:02 +08:00
Tianyi Cui c05e0953ef test(ptc): pin provider usage guidance in recorded program schemas 2026-09-12 15:03:02 +08:00
Tianyi Cui 2a08bf6ab9 feat(ptc): present provider execution guidance in logged program schema 2026-09-12 15:03:02 +08:00
Tianyi Cui c6b959783a test(ptc): record approved program execution and document per-call controls 2026-09-12 15:03:02 +08:00
Tianyi Cui a248cc4e64 feat(ptc): expose per-program timeout and sandbox approval controls 2026-09-12 15:03:02 +08:00
Tianyi Cui c5de252f8c docs(ptc): refresh persistence source references after master rebase 2026-09-12 15:03:02 +08:00
Tianyi Cui e21060defe test(code-runtime): observe native temp through inherited Windows handles 2026-09-12 15:03:02 +08:00
Tianyi Cui 091cbcdbfb fix(code-runtime): preserve native temp setup and preview activation 2026-09-12 15:03:02 +08:00
Tianyi Cui 2344830cfd test(code-runtime): align UI binding fixture and shared diagnostic ownership 2026-09-12 15:03:02 +08:00
Tianyi Cui bef3c26175 test(code-runtime): gate native enforcement cases on backend availability 2026-09-12 15:03:01 +08:00
Tianyi Cui 7f5eb6db9a fix(code-runtime): support SEA startup and immediate control writes 2026-09-12 15:03:01 +08:00
Tianyi Cui 98d3af0683 fix(code-runtime): separate model environment from native startup paths 2026-09-12 15:03:01 +08:00
Tianyi Cui a6f7eeadc6 fix(code-runtime): preserve native startup paths and preview packaging 2026-09-12 15:03:01 +08:00
Tianyi Cui ca7f3a523d test(tools): cover PTC standing policy and sandbox outcomes 2026-09-12 15:03:01 +08:00
Tianyi Cui 09e334622d docs(code-runtime): refresh confined runtime dependency graph 2026-09-12 15:03:01 +08:00
Tianyi Cui fc5cb0f649 test(code-runtime): complete sandboxed process coverage and catalogs 2026-09-12 15:03:01 +08:00
Tianyi Cui 8e19ec4962 feat(code-runtime): complete sandboxed Node execution and Session fixtures 2026-09-12 15:03:01 +08:00
Tianyi Cui f6904f4615 fix(tools): omit an absent Session during runtime policy resolution 2026-09-12 15:03:01 +08:00
Tianyi Cui 75ed8da3e0 feat(code-runtime): execute Node programs through confined processes 2026-09-12 15:03:01 +08:00
Tianyi Cui f95f7ec8dc refactor(code-runtime): name the Node runtime provider 2026-09-12 15:03:01 +08:00
Tianyi Cui b35a3b29eb fix(sandbox): pin ACL source resolution and recognize Node denials 2026-09-12 15:03:01 +08:00
Tianyi Cui e53c526320 test(subprocess): cover retained control endpoint teardown 2026-09-12 15:03:01 +08:00
Tianyi Cui 5bebc7c504 fix(sandbox): resolve source preload independently of command cwd 2026-09-12 15:03:01 +08:00
Tianyi Cui f5ef951ec5 test(subprocess): type rejected lifecycle results as unknown 2026-09-12 15:03:01 +08:00
Tianyi Cui 562e7f364e fix(subprocess): await control endpoint closure during disposal 2026-09-12 15:03:01 +08:00
Tianyi Cui 838da1d07e fix(subprocess): separate process quiescence from caller control streams 2026-09-12 15:03:01 +08:00
Tianyi Cui 96e12ddaf4 test(subprocess): cover source-only control launches and refresh generated references 2026-09-12 15:03:01 +08:00
Tianyi Cui c45d722d7b docs(subprocess): record the Windows duplex pipe requirement 2026-09-12 15:03:00 +08:00
Tianyi Cui 83a45679df fix(subprocess): create control carriers for overlapped Windows I/O 2026-09-12 15:03:00 +08:00
Tianyi Cui 63d462f28b test(subprocess): cover control disposal and runner forwarding 2026-09-12 15:03:00 +08:00
Tianyi Cui 1e5259c463 test(subprocess): type binary reads and native test spies precisely 2026-09-12 15:03:00 +08:00
Tianyi Cui 3ccb898cd5 test(subprocess): close control endpoint after the write callback 2026-09-12 15:03:00 +08:00
Tianyi Cui b9dfb8da1c test(subprocess): verify control inheritance and document channel ownership 2026-09-12 15:03:00 +08:00
Tianyi Cui d8efd4f8cd feat(subprocess): carry a duplex control pipe through managed launches 2026-09-12 15:03:00 +08:00
Tianyi Cui ff03531b97 docs(testing): record compaction fixture calibration 2026-09-12 15:02:37 +08:00
Tianyi Cui 65236780d1 test(ci): await final colors and build compactable history 2026-09-12 14:37:24 +08:00
turtle1999 0b368dfe78 test(persistence): budget heavy integration cases for CI 2026-09-12 14:35:03 +08:00
Tianyi Cui ee9ae1d0e0 docs(session): acknowledge Auto review error metadata 2026-09-12 14:22:33 +08:00
Tianyi Cui 2d9e3a175f docs(auto-review): name delegation records explicitly 2026-09-12 14:21:58 +08:00
Tianyi Cui 647e2aefbf Merge master into Auto review and reconcile recorded API types 2026-09-12 14:17:13 +08:00
Tianyi Cui 663a534d1f Merge pull request #3688 from deepseek-harness/policy-ci/policy-modules
refactor(ci): 分离 Issue policy 规则、GitHub IO 与生命周期
2026-09-12 13:31:13 +08:00
Tianyi Cui 0e77868739 Merge pull request #4032 from deepseek-harness/worktree/persistence-type-history
feat(session): enforce persistence type compatibility
2026-09-12 13:15:52 +08:00
Tianyi Cui 70a3be6f20 test(session): protect persistence digests from harmless reordering 2026-09-12 12:52:25 +08:00
Yifffan 1f76250952 Merge branch 'master' into fix/sidebar-document-preview-polish 2026-09-11 18:37:35 +02:00
Tianyi Cui f85b328450 chore(session): refresh persistence references after master merge 2026-09-12 00:13:33 +08:00
Yif 004dceed5e feat(sidebar-files): restore the tree's scroll position across tab switches 2026-09-12 00:07:50 +08:00
Yif adbe7e6395 fix(documentpreview): draw pdf page placeholders on the skeleton token 2026-09-12 00:07:40 +08:00
Tianyi Cui 7bbd7ef85c Merge commit 'cf74fa6ed7f3aad12b4e17e27be873d6b9438952' into worktree/persistence-type-history 2026-09-11 23:56:42 +08:00
Tianyi Cui 6e294b1c27 fix(session): infer persistence change decisions for agents 2026-09-11 23:56:06 +08:00
Yif 351b399639 feat(ui-commands): swap the feedback command icon to a paper plane 2026-09-11 23:42:32 +08:00
Yif f210305689 fix(documentpreview): fit images to the pane and polish binary and loading states 2026-09-11 23:42:20 +08:00
Turtle a0e1e11a86 Merge pull request #3996 from deepseek-harness/turtle/blacksmith-runtime-llm-e2e
ci: enable Blacksmith failover for runtime and DeepSeek e2e
2026-09-11 23:37:51 +08:00
Tianyi Cui 09a6fbc388 Merge pull request #4036 from deepseek-harness/worktree/ptc-sandbox/remove-e2b
refactor(e2b): retire the provider experiment and preserve remote seams
2026-09-11 23:30:29 +08:00
Turtle b625ebe440 Merge pull request #3151 from deepseek-harness/turtle/ban-ambiguous-origin-label
Ban an ambiguous origin label
2026-09-11 23:18:52 +08:00
Tianyi Cui a0a7e2c11c fix(session): harden and automate persistence change review 2026-09-11 23:15:26 +08:00
Tianyi Cui 845acdb3b7 refactor(ci): separate issue policy rules transport and lifecycle owners 2026-09-11 22:49:03 +08:00
Tianyi Cui f6e0d429e7 docs(ci): preserve current Issue label repair semantics 2026-09-11 22:48:20 +08:00
Tianyi Cui 2eeab3e65f fix(ci): preserve trusted legacy policy during preflight rollout 2026-09-11 22:42:46 +08:00
Tianyi Cui 108c87008b fix(ci): avoid unnecessary issue policy Project reads and lifecycle runs 2026-09-11 22:42:46 +08:00
Tianyi Cui 63a16e254d docs: clarify retained proxy and execution-world requirements 2026-09-11 22:35:08 +08:00
Tianyi Cui c49db8bc8c refactor(e2b): retire remote execution providers 2026-09-11 22:12:55 +08:00
Tianyi Cui 1e9b05d356 feat(session): verify persistence type history 2026-09-11 22:06:58 +08:00
lsdsjy 315e30e9f2 Merge pull request #3849 from deepseek-harness/turtle/issue-3848-headless-machine-readable-run
feat(headless): 支持 stdin 任务、--session-id 与 --json 事件流
2026-09-11 20:03:00 +08:00
lsdsjy fa5de9ea27 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 19:47:59 +08:00
Turtle 68f9708e02 fix(terminal-bash): bound retained storage and status reads 2026-09-11 19:12:10 +08:00
Turtle cea837e065 perf(terminal-bash): retain scrollback incrementally 2026-09-11 18:40:06 +08:00
Dudu-0223 eb56c23575 Merge pull request #3861 from deepseek-harness/fix/team-identity-context
fix(agent-team): preserve fork prefixes and use teammate-only direct delegation
2026-09-11 18:21:28 +08:00
Dudu-0223 950bdd3c4d Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 17:56:21 +08:00
imccyu 9edf90d33e Merge pull request #3984 from deepseek-harness/refactor/conversation-component-decomposition
refactor(conversation): extract panel and content components
2026-09-11 17:55:13 +08:00
Turtle 2b05a43eca fix: address concrete terminology review findings 2026-09-11 17:45:28 +08:00
Turtle 71d50b4a84 refactor: ban ambiguous origin label 2026-09-11 17:45:28 +08:00
Chinesezjc 62dd508cb0 test(permission): pin the disposed guard on the invalidation tick
The late-listener case now carries a pending generation change, so its
assertions depend on the disposed guard, and the silent-generation cases assert
their tick counts.
2026-09-11 17:45:23 +08:00
Dudu-0223 bfa6cc01aa fix(agent-team): raise default teammate limit to 16 2026-09-11 17:41:55 +08:00
imccyu 41915703cc refactor(conversation): move default views to its own module 2026-09-11 17:34:27 +08:00
imccyu f4dbec93df refactor(conversation): extract default views component 2026-09-11 17:34:27 +08:00
imccyu 8f5b1d71e3 refactor(conversation): move main panel to its own module 2026-09-11 17:34:27 +08:00
imccyu 0db88bd58d refactor(conversation): move content component to its own module 2026-09-11 17:34:26 +08:00
imccyu 2477554471 refactor(conversation): extract content component 2026-09-11 17:34:26 +08:00
imccyu 0220ec4989 refactor(conversation): extract main panel component 2026-09-11 17:34:26 +08:00
Dudu-0223 22b843ed2b Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 17:25:51 +08:00
Dudu-0223 f74d625e5a Merge pull request #4001 from deepseek-harness/fix/session-fork-boundary
fix(session): stop forks at the selected turn end
2026-09-11 17:25:35 +08:00
Dudu-0223 41de75a96d Merge pull request #3935 from deepseek-harness/fix/workspace-recency-order
fix(workspace): derive recency and pause the current order in Manual
2026-09-11 17:24:49 +08:00
Chinesezjc eb6373d711 fix(permission): tick only on a real generation change
A notification repeating the published generation no longer withdraws displayed
options, so the invalidation observable and the README describe the same rule.
The composer seat hides on a cleared catalog while the slash picker keeps its
failure and retry, and the disposed guard that both callers already provide is
gone from the tick.
2026-09-11 17:20:22 +08:00
Dudu-0223 589cc7ae87 test(web): align Team fixture with shipped delegation tools 2026-09-11 17:18:58 +08:00
Dudu-0223 dafff8e51c fix(subprocess): settle consumed empty scopes before exit notification 2026-09-11 17:05:15 +08:00
Chinesezjc 38391bf6ba Merge remote-tracking branch 'origin/master' into auto-review-per-call 2026-09-11 17:03:09 +08:00
Chinesezjc 14bdce3644 docs(permission): describe dismissal as a catalog invalidation
The package README, its Chinese pair, and the catalog spec now state both
public observables and pin the generation-change tick that withdraws displayed
options.
2026-09-11 17:00:54 +08:00
Dudu-0223 ca53514562 fix(ci): use HTTPS Ubuntu archives on Blacksmith 2026-09-11 17:00:30 +08:00
Dudu-0223 a68d3131c8 Merge branch 'master' into fix/session-fork-boundary 2026-09-11 16:56:46 +08:00
Dudu-0223 5731647698 test(agent-team): follow the DeepSeek protocol module move 2026-09-11 16:48:15 +08:00
Dudu-0223 96a3a32fa2 fix(ci): use HTTPS Ubuntu archives on Blacksmith 2026-09-11 16:47:31 +08:00
Dudu-0223 8eafb70b04 fix(ci): use HTTPS Ubuntu archives on Blacksmith 2026-09-11 16:47:09 +08:00
Dudu-0223 10175e62cc Merge master into fix/team-identity-context 2026-09-11 16:46:22 +08:00
Dudu-0223 ad3be9711d Merge master into fix/workspace-recency-order 2026-09-11 16:46:21 +08:00
lsdsjy c58ddc5fa3 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 16:41:56 +08:00
Yichen Jiang 7485d75e38 Merge pull request #3682 from deepseek-harness/worktree/llm-deepseek-messages
feat(llm): add Messages protocol support to DeepSeek
2026-09-11 16:38:43 +08:00
lsdsjy 62c04f12cc Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 16:37:19 +08:00
Dudu-0223 4984befd2c Merge master into fix/team-identity-context 2026-09-11 16:32:32 +08:00
Dudu-0223 b6e9b013f9 Merge master into fix/workspace-recency-order 2026-09-11 16:32:31 +08:00
Chinesezjc 4a935ed8fd fix(permission): drop the picker only on a catalog invalidation
Publishing the result of a read the picker waits for no longer closes that
picker, so a failed read keeps its error and retry state; a catalog
notification or generation change still withdraws stale options. Correct the
action-history rationale in the Auto review note and cover both paths with the
real popup controller.
2026-09-11 16:29:31 +08:00
Chinesezjc e995d71a0e fix(permission): keep the picker retryable after a failed catalog read
Availability now follows the Session projection alone, so a failed catalog
read surfaces through the picker's own retry instead of hiding the command.
Type-guard the persisted denial reason before trimming it, and record in the
Auto review note why the reviewer reads the complete action history.
2026-09-11 15:46:19 +08:00
Yichen Jiang bc13dcf550 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-11 15:42:51 +08:00
Chinesezjc 1de37ddc2b test(web): follow the settled-loader startup audit
Loader settlement no longer rejects on a failed optional entry, so the Auto
rollback scenario reads the conflict from the startup audit. Refresh the Auto
denial golden for the merged composer chrome.
2026-09-11 15:41:46 +08:00
creatixchu 93581008aa docs: refresh the paired Session API record 2026-09-11 15:40:44 +08:00
Dudu-0223 4f1663c143 Merge remote-tracking branch 'origin/master' into fix/session-fork-boundary 2026-09-11 15:39:29 +08:00
Dudu-0223 88c3d05fbf fix(session): align browser fixture forks and legal log regressions 2026-09-11 15:39:20 +08:00
creatixchu 653a4164b8 docs: sync the projection disposer signature 2026-09-11 15:38:19 +08:00
creatixchu 4e7193945d fix: expose the projection registration async disposer 2026-09-11 15:38:04 +08:00
creatixchu 325c8d880a build: admit pure message projection leaves in browser bundles 2026-09-11 15:35:55 +08:00
Turtle 4c4b7e6cb2 Merge pull request #4003 from deepseek-harness/turtle/systemd-scope-exit-race
fix(ci): stabilize teardown, process sampling, and reply queue coverage
2026-09-11 15:35:08 +08:00
creatixchu 401c3ef977 test: type the stored projection fixture as format events 2026-09-11 15:29:33 +08:00
Dudu-0223 b16687ffcb Merge branch 'master' into fix/team-identity-context 2026-09-11 15:27:52 +08:00
creatixchu 1d4d22c1ff test: complete projection fixtures and configuration catalog 2026-09-11 15:26:54 +08:00
Dudu-0223 75b89b353d Merge branch 'master' into fix/workspace-recency-order 2026-09-11 15:25:54 +08:00
creatixchu 413ac14b16 refactor: move image offload projection into its plugin 2026-09-11 15:24:45 +08:00
Dudu-0223 11248ea949 fix(workspace): retain blank manual positions during reconnect 2026-09-11 15:21:09 +08:00
Dudu-0223 973bea8204 fix(session): use the selected turn end as the fork cut 2026-09-11 15:09:47 +08:00
Chinesezjc 6b19923ad5 fix(auto-review): clear the master integration gates
Register the permission catalog default response so the web roster boot
matches every Remote call it makes, and carry the deprecation rule for the
session history read this reviewer needs with the deferred migration
recorded in the package README.
2026-09-11 15:08:48 +08:00
Turtle 16f80298df Merge pull request #4006 from deepseek-harness/turtle/boot-failure-e2e
test(boot): cover Web failure recovery and inject module routes
2026-09-11 15:04:29 +08:00
Dudu-0223 0f8552df7f fix(session): retain fork tail until the first inbox change 2026-09-11 14:58:46 +08:00
turtle1999 fd0dce2f20 ci: route runtime and LLM e2e through Blacksmith failover 2026-09-11 14:51:15 +08:00
Dudu-0223 41196ae591 fix(agent-team): disable direct subagent tools in Team profiles 2026-09-11 14:50:52 +08:00
turtle1999 18a1955b7e fix(ci): handle scope termination races and cyclic process snapshots 2026-09-11 14:50:46 +08:00
turtle1999 85c37d1950 docs(boot): document startup and reload failure behavior 2026-09-11 14:45:37 +08:00
Yichen Jiang a326698ee7 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages
# Conflicts:
#	docs/config-catalog.i18n.yaml
#	docs/config-catalog.md
#	docs/config-catalog.zh.md
#	packages/llm/llm-deepseek/src/adapter.ts
#	packages/llm/llm-deepseek/src/common/request-pricing.ts
#	packages/llm/llm-deepseek/src/index.ts
#	packages/llm/llm-deepseek/tests/adapter.spec.ts
#	packages/llm/llm-deepseek/tests/image-tokens.spec.ts
2026-09-11 14:44:30 +08:00
turtle1999 eb9da8930e test(boot): cover Web failure recovery and inject module routes 2026-09-11 14:34:17 +08:00
Dudu-0223 3a48b62344 refactor(workspace): derive flat rows only after ordering 2026-09-11 14:31:25 +08:00
Chinesezjc 3afad043c6 Merge remote-tracking branch 'origin/auto-review-per-call' into auto-review-per-call 2026-09-11 14:29:11 +08:00
Chinesezjc 358d0bc49c Merge remote-tracking branch 'origin/master' into auto-review-per-call 2026-09-11 14:27:30 +08:00
Chinesezjc d8c54899b7 fix(experimental): align the Auto review package version with the release 2026-09-11 14:27:27 +08:00
creatixchu bc49b2a15c Merge master 28a4010f7d into image offload branch 2026-09-11 14:22:36 +08:00
_Kerman 20106ecda8 Merge pull request #3869 from deepseek-harness/xtr/ptc-sdk-literal-prompt
fix(ptc): preserve literal braces in generated SDK prompts
2026-09-11 14:20:23 +08:00
Chinesezjc a605cf579d Merge branch 'master' into auto-review-per-call 2026-09-11 14:19:16 +08:00
Dudu-0223 896a6c3e47 fix(session): stop forks at the selected turn end 2026-09-11 14:17:53 +08:00
Chinesezjc fed65b7cad Merge master into the Auto review branch
Resolve 36 conflicts across source, bilingual READMEs, and generated
documentation: keep master's command-identity and popup APIs, RemoteMock
test infrastructure, and regenerated catalogs; keep the Auto review
catalog, slot, dismissal, and snapshot content.
2026-09-11 14:15:32 +08:00
Yifffan ef39ebd0bb Merge pull request #3969 from deepseek-harness/refine/connecting-ui
feat(sidebar): refine the connection indicator states and interaction
2026-09-11 08:06:56 +02:00
Dudu-0223 d19d23cf86 fix(workspace): make manual session order browser-local 2026-09-11 14:05:13 +08:00
_Kerman 1d20d46a08 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3869 2026-09-11 13:54:03 +08:00
Turtle ef0041b351 Merge pull request #3964 from deepseek-harness/turtle/master-ci-browser-realpath
fix(ci): repair browser roots and asynchronous test fixtures
2026-09-11 13:49:06 +08:00
creatixchu c59d285f99 chore(compaction): mark existing history read for deferred migration 2026-09-11 13:43:02 +08:00
creatixchu ec3560c42e Merge master 16bf879af2 into image offload branch 2026-09-11 13:40:19 +08:00
lsdsjy 05373a4897 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-11 13:37:34 +08:00
Yifffan c160884c19 Merge branch 'master' into refine/connecting-ui 2026-09-11 07:32:49 +02:00
CreatixChu ac34c5deac Merge pull request #3938 from deepseek-harness/worktree/issue-3929-v41-request-image
更新v41的图片大小限制
2026-09-11 13:32:15 +08:00
turtle1999 40e39f4673 test(ci): control Python stray UTF-8 fragment delivery 2026-09-11 13:31:22 +08:00
Turtle 239f695e1f Merge pull request #3830 from deepseek-harness/turtle/required-plugin-startup-policy
feat(boot): distinguish required startup failures
2026-09-11 13:28:59 +08:00
Yif ce1a55816f Merge remote-tracking branch 'origin/refine/connecting-ui' into refine/connecting-ui 2026-09-11 13:28:50 +08:00
turtle1999 96ee6e49f9 test(terminal): await PowerShell prompt delivery 2026-09-11 13:22:55 +08:00
Yichen Jiang e5e9037dfb test(llm): opt in Messages system-update e2e 2026-09-11 13:17:25 +08:00
creatixchu db6dff8f67 Merge remote-tracking branch 'origin/master' into worktree/issue-3929-v41-request-image 2026-09-11 13:16:24 +08:00
turtle1999 464334bc4c test(boot): keep headless running after optional MCP startup failure 2026-09-11 13:10:03 +08:00
Turtle b576419a4c Merge branch 'master' into turtle/master-ci-browser-realpath 2026-09-11 13:03:10 +08:00
Turtle f93cea5734 Merge pull request #3983 from deepseek-harness/turtle/master-ci-windows-store
fix(ci): derive Windows standby store from workspace volume
2026-09-11 13:01:36 +08:00
_Kerman 95690d6453 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3869 2026-09-11 12:55:09 +08:00
Yichen Jiang 6ad0ddb899 test(llm): pin Messages e2e to its official endpoint 2026-09-11 12:51:25 +08:00
turtle1999 6894e93a12 refactor(app-boot): keep required startup policy private 2026-09-11 12:47:48 +08:00
turtle1999 4cd20c1a1e fix(app-boot): bound cyclic startup error diagnostics 2026-09-11 12:47:48 +08:00
turtle1999 25d5efcd86 Classify disabled expression errors in startup audits 2026-09-11 12:47:48 +08:00
turtle1999 f5b5a9da51 Require Web modules and connection during startup 2026-09-11 12:47:47 +08:00
turtle1999 97d1f7c203 fix(web-app): gate readiness on required startup entries 2026-09-11 12:47:47 +08:00
turtle1999 7fd41b833b fix(boot): limit required IDs to application endpoints 2026-09-11 12:47:47 +08:00
turtle1999 f4a7dd6077 fix(boot): preserve failure policy on nontransactional Loader 2026-09-11 12:47:46 +08:00
Turtle f308742050 test(boot): cover startup diagnostic variants 2026-09-11 12:47:46 +08:00
Turtle 941267b197 test(boot): type fixture fiber state 2026-09-11 12:47:46 +08:00
Turtle fbcad258ad test(boot): keep required fixture failures fatal 2026-09-11 12:47:46 +08:00
Turtle bd4cfc7c46 feat(boot): distinguish required startup failures 2026-09-11 12:47:46 +08:00
Yif dd7ba51a13 Merge remote-tracking branch 'origin/master' into refine/connecting-ui
# Conflicts:
#	packages/client/ui-primitives/README.i18n.yaml
#	packages/client/ui-primitives/README.md
#	packages/client/ui-primitives/README.zh.md
2026-09-11 12:29:34 +08:00
Ziya 1f56a2c0e0 Merge pull request #3970 from deepseek-harness/copy/ask-user-skip-20260910
fix(web): 简化提问卡片的中英文跳过文案
2026-09-11 11:54:51 +08:00
ZiyaZhang 91da8a4ecf Merge master and retain its Python coverage fix 2026-09-10 20:38:00 -07:00
Yichen Jiang bc99474e87 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages
# Conflicts:
#	packages/llm/llm-deepseek/package.json
2026-09-11 11:27:47 +08:00
creatixchu f0f1988a7a Clarify token counts after aspect-preserving image projection 2026-09-11 11:14:33 +08:00
creatixchu 08dba5f5f6 test(compaction): record summary image recovery through ACP 2026-09-11 11:00:17 +08:00
creatixchu 73e10fb80c fix(compaction): preserve summary diagnostics on cancellation 2026-09-11 10:58:36 +08:00
creatixchu debb4b9a9c fix(compaction): recover durable image offload in summary requests 2026-09-11 10:56:48 +08:00
creatixchu 275af6a2e2 Merge remote-tracking branch 'origin/master' into worktree/issue-3929-v41-request-image 2026-09-11 10:42:03 +08:00
turtle1999 e46cb49103 fix(ci): derive Windows standby store from workspace volume 2026-09-10 23:02:21 +08:00
Yichen Jiang c291e7961a Merge pull request #3977 from deepseek-harness/worktree/release-0.1.5-sync-master
feat(web): sync feedback and file refinements from release
2026-09-10 22:17:09 +08:00
Turtle e570e74705 Merge pull request #3962 from deepseek-harness/turtle/ci-smoke-20260910-1820
test(python): make stray-output sealing coverage deterministic
2026-09-10 22:01:34 +08:00
Yichen Jiang 59f2e3be33 Merge release/dsh-0.1.5 version 0.1.5-rc.2 into master sync branch 2026-09-10 21:53:53 +08:00
Yichen Jiang fb2c4b9e69 Merge pull request #3978 from deepseek-harness/worktree/release-dsh-0.1.5-rc.2
release(dsh): 0.1.5-rc.2
2026-09-10 21:50:19 +08:00
Yichen Jiang a305303422 release(dsh): 0.1.5-rc.2 2026-09-10 21:38:20 +08:00
Dudu-0223 a17f0733b1 test(workspace): stabilize sidebar captures and update blank placement 2026-09-10 21:34:45 +08:00
Dudu-0223 4857614be0 fix(workspace): pause current ordering in manual mode 2026-09-10 21:25:09 +08:00
Yichen Jiang 3bfcebe15e Merge master into worktree/release-0.1.5-sync-master 2026-09-10 21:23:11 +08:00
Yichen Jiang 2107e4697e Merge pull request #3973 from deepseek-harness/worktree/release-0.1.5-backport-feedback-deliverables
feat(web): backport feedback and file refinements to 0.1.5
2026-09-10 21:09:45 +08:00
Yifffan 879949e1c9 Merge branch 'master' into refine/connecting-ui 2026-09-10 14:55:03 +02:00
yixiangihsiangandLegGasai 060323d8e2 feat(web): backport feedback and file refinements to 0.1.5
Backport the selected changes from PR #3928 and PR #3902 onto dsh-v0.1.5-rc.1, retaining baseline session recordings and excluding newer master commits.

Refs #3927
Refs #3903

Co-authored-by: LegGasai <[email protected]>
2026-09-10 20:51:46 +08:00
ZiyaZhang 884485e7e6 test(python): restore concat through its spy 2026-09-10 05:40:50 -07:00
ZiyaZhang 7bce0c26f5 test(python): make stray fragment coverage deterministic 2026-09-10 05:35:32 -07:00
Yif 11ca8020cd fix(sidebar): keep one reconnect label and confirm recovery from visibility
ConnectionController deduplicates repeated connecting states across backoff
attempts, so the shell cannot tell manual retries from automatic ones; a
shell-held flag either flipped the label mid-hold or stuck across later
attempts. Every attempt now reads the single connecting label, and the
two-second recovery confirmation starts when the recovered pill becomes
visible rather than at the state transition, so the 800ms hold never
shortens it.
2026-09-10 20:12:09 +08:00
ZiyaZhang 69db89f848 fix(web): shorten English question skip label 2026-09-10 05:06:14 -07:00
ZiyaZhang 2eed93628c fix(web): shorten Chinese question skip label 2026-09-10 05:00:38 -07:00
Yif b044e45438 test(web): pin the static connection pill in the lifecycle browser snapshot 2026-09-10 19:52:38 +08:00
lsdsjy 71b2cc80aa fix(headless): waive the deferred Session history read
Master's `deprecate-synchronous-session-event-reads` marks `Session.eventAt`
deprecated and rejects new production calls. `liveEvents` predates that decision
and still needs the ordered log to reconstruct the preset a session runs under,
so carry the repository's standard line-scoped waiver for deferred migration.
2026-09-10 19:32:54 +08:00
Yif 16e986548e test(sidebar): pin the refined connection indicator behavior and docs
Update the behavior tests for the static retry label, the 800ms minimum
connecting hold, manual-versus-automatic attempt naming, and the 150ms
fade-out unmount; restate the interaction in both packages' bilingual
READMEs and record the decision as an Agent Note.
2026-09-10 19:29:48 +08:00
lsdsjy 79bd986fa7 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-10 19:16:53 +08:00
lsdsjy 145f20f712 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-10 19:13:38 +08:00
lsdsjy 3be1ea10b9 refactor(headless): move the runner process hooks out of the package entry
Address the Code Entropy follow-up on d9c4f53758: the runner entry still
expanded its process-substitution object with a `readStdin` hook used only by
tests. Move that seam to `src/runner-internals.ts`, mirroring
`src/startup-internals.ts`, so neither entry grows package API for tests.
2026-09-10 19:13:05 +08:00
turtle1999 464e1cce49 fix(ci): canonicalize browser dependency scan roots 2026-09-10 19:01:00 +08:00
turtle1999 0963bdd7d3 test(python): control stray-output fragments for sealing coverage 2026-09-10 18:29:04 +08:00
turtle1999 39e6e6fb89 Retain awaited client cleanup after rebase 2026-09-10 18:21:19 +08:00
turtle1999 0e6505454e Align client boot import failure test with reverted Loader 2026-09-10 18:18:10 +08:00
turtle1999 ab4d22595b Merge master and await client plugin cleanup after removal 2026-09-10 18:18:10 +08:00
turtle1999 3b5daf10be test(ci): verify current master on PR runners 2026-09-10 18:16:47 +08:00
Turtle 060ae6f336 Merge pull request #3957 from deepseek-harness/fix/subprocess-linux-scope-empty-range
fix(ci): resolve Blacksmith scope teardown and hosted test failures
2026-09-10 18:11:48 +08:00
_Kerman 42b50bd377 Merge pull request #3828 from deepseek-harness/xtr/deprecate-session-event-readers
docs(session): deprecate synchronous event-history reads
2026-09-10 18:03:54 +08:00
Chinesezjc 1050091d82 Merge pull request #3926 from deepseek-harness/ci/blacksmith-hosted-image-fixes
test/ci: remove the host-paced assumptions the hosted coverage lanes hit
2026-09-10 18:01:54 +08:00
turtle1999 989b3d4390 Merge hosted-image test fixes for Blacksmith CI 2026-09-10 17:53:36 +08:00
07akioni 69005cb956 Merge pull request #3860 from deepseek-harness/fix/parallel-macos-notarization
perf(desktop): 并行 macOS 公证并记录上传加速说明
2026-09-10 17:52:40 +08:00
Chinesezjc 0377049465 ci: give the Linux coverage lane the hosted image's teardown budget
The managed-scope disposal cases in subprocess-local and bash-sandbox run past
the 5000ms default when this lane's four partitions, six workers per gate, and
sibling gates share one host. The same commit stays inside the budget on the
in-house pool: the hosted image exceeded it (run 34449848541) and the lane is
green with 90000 (run 34457655892). The Windows coverage lane already grants
the same budget for the same class of teardown.

(cherry picked from commit bf3468adf0947a51f09d4a1aaae2865f3e50a2f2)
2026-09-10 17:47:27 +08:00
Chinesezjc 114e8c9b09 test: drop two host-paced assumptions the hosted image exposes
The ACP plugin-config case drove a real dispose with 150ms graces. On the hosted
image the SIGKILL escalation landed while the managed scope could not take the
signal — systemctl failed the kill ("Failed to send signal SIGKILL to auxiliary
processes: Invalid argument") — so the run reported a teardown failure the
configuration never asked for. The case asserts that config graces reach the
real run, so the values only need to clear the host's scope handling. Its mock
refuses stdin EOF and SIGTERM by design, so the case waits out both graces and
carries its own budget for that fixed cost.

Both illegal-UTF-8 residual cases relied on a plain sched_yield leaving each
write as its own read chunk. A starved reader coalesces hundreds of bytes into
one chunk, which inflates the measured peak for correct accounting too: the
hosted image measured 2563 against the 2048 bound. Pace with a real sleep so
each byte is its own chunk everywhere, and keep each payload above the bound
(3200 bytes; 1100 CESU-8 sequences) so a raw-byte undercount still flushes past
it instead of only at EOF.

(cherry picked from commit 30e0c3ff0415befe3652a55611df8a18d82c18bb)
2026-09-10 17:47:27 +08:00
Chinesezjc c4c2d16308 test(subprocess): align the teardown race with the settlement contract
`disposal contains a spawn-failure rejection that races teardown` asserted one
winner of the race. A bootstrap that publishes its pre-exec failure rejects with
that failure, and a teardown that stops the bootstrap first settles as the
requested termination — the recorded failure only outranks the stop when it was
published before the stop landed. Assert the contract instead: a rejection
carries the failure, a stop carries SIGTERM, which only the Linux scope records
because the win32 job owner and the fallback launcher both reject the start.

The two terminal cases that drive a mocked PTY exit state why they pin
`internals.platform`: under the host's native scope the mocked exit races the
scope bootstrap and fails as `terminal scope exited before its bootstrap
consumed the launch request`.

(cherry picked from commit fee113eb9a5632555304858d2bf6355588a3aaf0)
2026-09-10 17:47:27 +08:00
Yif 626fa816a5 feat(sidebar): refine the connection indicator states and styling
The disconnected pill now names the retry action statically (refresh icon +
"连接异常,刷新重试") instead of swapping text on hover; connecting shows a
spinner, holds visible for at least 800ms, and distinguishes manual retry
(重新连接中) from automatic backoff (自动重连中); state changes fade over
150ms. Pill chrome tightened to 28px height with a 20%-alpha label border.
2026-09-10 17:45:56 +08:00
Turtle def3f88bf5 Merge pull request #3943 from deepseek-harness/fix/master-ci-corpus-css-pin
fix(ci): admit any `.css` refusal for the dockkit corpus exemption
2026-09-10 17:44:34 +08:00
creatixchu ffa536712a Merge remote-tracking branch 'origin/master' into worktree/issue-3929-v41-request-image
# Conflicts:
#	packages/attachment/attachment-local/README.i18n.yaml
#	packages/attachment/attachment-local/README.md
#	packages/attachment/attachment-local/README.zh.md
2026-09-10 17:41:17 +08:00
Turtle 3ece27847c Merge pull request #3925 from deepseek-harness/turtle/dsh-cache-path
feat(attachment-local): use shared cache for request images
2026-09-10 17:38:26 +08:00
Chinesezjc 87f32dc8ca docs(notes): record the hosted-image assumptions the coverage suite names
The failover leg runs this suite on pools this repository does not own, and the
coverage lanes failed there on host properties their cases never named: the
containment a mocked PTY exit raced, the grace a managed scope needs before it
can take a SIGKILL, reader coalescing in the illegal-UTF-8 residual cases, and
a Windows image that refuses the folder dialog class. Record what each case
pins now, what was rejected, and the runs that measured it, so the next pool
change starts from those host dependencies instead of re-diagnosing them.
2026-09-10 17:36:16 +08:00
Chinesezjc 45c10ed2ce ci: give the Linux coverage lane the hosted image's teardown budget
The managed-scope disposal cases in subprocess-local and bash-sandbox run past
the 5000ms default when this lane's four partitions, six workers per gate, and
sibling gates share one host. The same commit stays inside the budget on the
in-house pool: the hosted image exceeded it (run 34449848541) and the lane is
green with 90000 (run 34457655892). The Windows coverage lane already grants
the same budget for the same class of teardown.
2026-09-10 17:36:16 +08:00
Chinesezjc f45ca2d9ac test: drop two host-paced assumptions the hosted image exposes
The ACP plugin-config case drove a real dispose with 150ms graces. On the hosted
image the SIGKILL escalation landed while the managed scope could not take the
signal — systemctl failed the kill ("Failed to send signal SIGKILL to auxiliary
processes: Invalid argument") — so the run reported a teardown failure the
configuration never asked for. The case asserts that config graces reach the
real run, so the values only need to clear the host's scope handling. Its mock
refuses stdin EOF and SIGTERM by design, so the case waits out both graces and
carries its own budget for that fixed cost.

Both illegal-UTF-8 residual cases relied on a plain sched_yield leaving each
write as its own read chunk. A starved reader coalesces hundreds of bytes into
one chunk, which inflates the measured peak for correct accounting too: the
hosted image measured 2563 against the 2048 bound. Pace with a real sleep so
each byte is its own chunk everywhere, and keep each payload above the bound
(3200 bytes; 1100 CESU-8 sequences) so a raw-byte undercount still flushes past
it instead of only at EOF.
2026-09-10 17:36:15 +08:00
Chinesezjc df709f442c test(subprocess): align the teardown race with the settlement contract
`disposal contains a spawn-failure rejection that races teardown` asserted one
winner of the race. A bootstrap that publishes its pre-exec failure rejects with
that failure, and a teardown that stops the bootstrap first settles as the
requested termination — the recorded failure only outranks the stop when it was
published before the stop landed. Assert the contract instead: a rejection
carries the failure, a stop carries SIGTERM, which only the Linux scope records
because the win32 job owner and the fallback launcher both reject the start.

The two terminal cases that drive a mocked PTY exit state why they pin
`internals.platform`: under the host's native scope the mocked exit races the
scope bootstrap and fails as `terminal scope exited before its bootstrap
consumed the launch request`.
2026-09-10 17:36:14 +08:00
Chinesezjc ac5f851095 test(directory-picker): probe the folder dialog class instead of the platform
A Windows Server image can answer CoCreateInstance(CLSID_FileOpenDialog)
with CLASS_E_CLASSNOTAVAILABLE, so the win32 COM smoke has no dialog to
open and the POSIX rejection expectation is the only proof that image can
produce. Gate the pair on a PowerShell activation probe taken outside the
code under test, so a GUID or vtable regression in src still fails the
smoke instead of skipping it.
2026-09-10 17:36:13 +08:00
lsdsjy 40dc25fe1d Merge remote-tracking branch 'origin/master' into HEAD 2026-09-10 17:32:26 +08:00
lsdsjy 6d05765405 refactor(headless): keep the startup process seam out of the public entry
Address the Code Entropy finding on f67d722243: the `./startup` entry exported
`internals` only so tests could substitute process facts, which added avoidable
package API. Move the seam to `src/startup-internals.ts`, which no package
export names, and have the provider and its tests import it from there.
2026-09-10 17:29:15 +08:00
turtle1999 aaa02a3970 fix(subprocess): settle a Linux scope left active with no processes
Range observation waited for an active transient scope to leave `active`
before it would prove the managed range empty. systemd ends a scope only when
it observes the populated-to-empty transition, so a payload killed before the
manager placed it in the scope's cgroup leaves the unit active with an empty
cgroup forever. `waitForExit()` then polled a unit that could not change, and
`LocalSubprocessRuntime.disposeManagedProcesses()` awaited it without bound —
teardown hung for any command cancelled immediately after spawn.

Each state query now also reads the unit's process count. Once termination was
requested and the launcher has exited, a still-active unit reporting no
processes is the empty range: the launcher forked every process it will ever
fork, so nothing later joins that cgroup. The owner stops the leftover unit so
transient units cannot accumulate, and an unreported or `[not set]` count
stays unknown and keeps waiting.

The race needs a live user-systemd manager, so it stayed dormant wherever the
native scope path was unavailable and surfaced once CI routed lanes onto the
Blacksmith hosted image, whose runners accept that path. Ten repetitions of
`local.spec.ts > disposal kills still-running processes and awaits their exit`
in a Linux container with a user manager failed seven times before the change
and passed ten times after it.
2026-09-10 17:27:52 +08:00
Yichen Jiang 59c1ee7c82 test(web): retain historical Messages replay after provider consolidation 2026-09-10 17:27:32 +08:00
creatixchu 4b1408e1df Merge origin/master into worktree/3041-image-offload-log-design 2026-09-10 17:20:54 +08:00
Yichen Jiang 15dd4fa046 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-10 17:20:22 +08:00
Yichen Jiang 6a137ea702 refactor(llm): unify DeepSeek protocol implementations 2026-09-10 17:19:51 +08:00
_Kerman 0989a5cffc Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3828 2026-09-10 17:17:09 +08:00
07akioni 42123ef019 fix(desktop): tighten parallel notarization checks 2026-09-10 17:12:45 +08:00
_Kerman 5a022a0db1 Merge latest origin/master into xtr/agent-setup-hook 2026-09-10 17:04:05 +08:00
_Kerman fd7ee32315 Merge origin/master into xtr/agent-setup-hook 2026-09-10 17:02:57 +08:00
07akioni ce94430e16 Merge master into fix/parallel-macos-notarization 2026-09-10 17:00:35 +08:00
imccyu 25d554d46e Merge pull request #3650 from deepseek-harness/worktree-feqatest
feat(test-support): whole-client test tier over an endpoint-named Remote mock
2026-09-10 16:58:13 +08:00
_Kerman 8af1cb79f4 docs(core): update session start source event reference 2026-09-10 16:54:52 +08:00
07akioni 6fc0064b70 Merge pull request #3914 from deepseek-harness/fix/bundle-speed
perf(desktop): 优化运行时打包、启动及错误恢复
2026-09-10 16:54:32 +08:00
lsdsjy 9b330b653f docs(headless): correct the adopt-only wording and lockfile attribution
Address the ds-review-bot suggestions on 4099b84ee8:

- The post-resume comment scopes the write lease to other processes, so it no
  longer contradicts the post-idle re-check that defends against a same-process
  overlay writer.
- The Agent Note attributes `pnpm-lock.yaml` to the new
  `@deepseek-ai/dsh-session-query` workspace link instead of the shared chunk,
  which `scripts/check-workspace-constraints.ts` and the manifest publish.
- The package summary keeps "leaves nothing running behind", so it does not
  imply the runner leaves no persisted Session behind.
2026-09-10 16:52:43 +08:00
creatixchu 31f731cc6f Merge remote-tracking branch 'origin/master' into codex/pr-3440-resync-20260908
# Conflicts:
#	docs/subsystems/llm-streaming.i18n.yaml
#	docs/subsystems/llm-streaming.md
#	docs/subsystems/llm-streaming.zh.md
#	packages/llm/llm-deepseek/README.i18n.yaml
#	packages/llm/llm-deepseek/README.md
#	packages/llm/llm-deepseek/README.zh.md
#	packages/llm/llm-deepseek/src/request-pricing.ts
#	packages/llm/llm-deepseek/tests/request-pricing.spec.ts
2026-09-10 16:52:18 +08:00
turtle1999 5a88e5c4b9 docs(ci): state the full admitted failure class for the dockkit exemption
The package README and the Agent Note named an extension or error code but
omitted the unrelated-message case the classifier also rejects, and the note
loosened "Node's unknown-`.css`-extension refusal" to any non-`.css` failure.
2026-09-10 16:50:43 +08:00
creatixchu b5e7fca4a5 feat(compaction): record image offload decisions without message replacement 2026-09-10 16:47:50 +08:00
creatixchu ba30b73f7b refactor(attachment): 求解器留在 llm-deepseek,存储层只接收目标尺寸
ImageRequestPolicy 改为 ImageRequestTarget,只含目标宽、高和字节目标;dsh-attachment 删除 token-grid 投影联合和 tokenGridProjection,只保留提供方无关的 requestImageDimensions 与 longEdgeDimensions。v41 求解器回到 llm-deepseek 的 image-tokens.ts,新增 deepSeekRequestImageDimensions,resolveRequestImageTarget 为每张图算出目标并加 4096 单边上限;pi-ai 路由按自身像素预算推导目标。

按 review 意见修正:v41 计价 Note 里已删除常量的过期句子就地改写;ProjectedDimensions 登记到 attachment 子系统页与 type-equiv 清单;Note 区分发送尺寸与提供方补齐后的网格;短边取整的措辞改为「按路由预测」。
2026-09-10 16:46:28 +08:00
_Kerman 632d42ee11 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3828 2026-09-10 16:42:45 +08:00
lsdsjy ef34e9d888 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-10 16:42:31 +08:00
imccyu f74ca1e266 fix(remote-mock): abort stream handle on consumer return 2026-09-10 16:34:32 +08:00
imccyu 4ffc774e5c test(settings): exercise settings through the whole-client fixture 2026-09-10 16:34:31 +08:00
imccyu d58c5455dd test(api): migrate Session and Workspace specs to assembled clients 2026-09-10 16:34:30 +08:00
imccyu 627c0f741e feat(test-support): assemble real clients with native test fixtures 2026-09-10 16:34:29 +08:00
imccyu 48d83eca66 refactor(client): share boot and transport integration points 2026-09-10 16:34:28 +08:00
imccyu 3d6fa12a6d feat(remote-mock): expose typed native mocks and controlled streams 2026-09-10 16:34:27 +08:00
imccyu 0718711886 docs(testing): define the native-mock client assembly tier 2026-09-10 16:34:27 +08:00
CreatixChu 8f85739d7b Merge pull request #3745 from deepseek-harness/worktree/composer-plus-menu
optimize: inputbar slash group and locale
2026-09-10 16:31:39 +08:00
turtle1999 529434e1df fix(ci): admit any .css refusal for the dockkit corpus exemption
The built-bundle import sweep pinned the Dockkit exemption to
`packages/client/ui-dockkit/lib/components/dockkit.module.css`, but the built
bundle imports `@deepseek-ai/dsh-client-ui-primitives` before its own
stylesheet, and the tsx launcher resolves that specifier through tsconfig
`paths` into the dependency's `src` tree. The sweep therefore reports
`packages/client/ui-primitives/src/StateDot.module.css`, so the pinned path can
never match where client build output exists; the Windows complete-gate
inventory reports the exempt bundle as an unexpected baseline failure.

Admit Node's unknown-`.css`-extension refusal for any stylesheet and keep every
other failure a finding. The classification controls cover an admitted Dockkit
stylesheet, the dependency's source stylesheet, another extension, an arbitrary
message, another error code, and a stale exemption.

Agent Note: .agents/notes/implemented/bug-fix/2026-09-10-built-bundle-css-exemption.md
2026-09-10 16:28:45 +08:00
lsdsjy 31be030ccc fix(headless): keep --json stderr free of commander's error print
Address the ds-review-bot findings on 8724223621:

- The JSON error override throws commander's control-flow error instead of
  delegating to `program.error`, whose own printer wrote an unprefixed
  `error: …` line to stderr and broke the documented `--json` contract.
- Cover the post-idle adoptability re-check with a focused case: an overlay that
  appends a preset while the runner awaits idle is still rejected, and the stale
  "live Session" comment now names that residual writer.
- Scope the Agent Note change sentence to the product change so it no longer
  claims the whole patch is confined to one package.
2026-09-10 16:25:19 +08:00
_Kerman 0600fa405b test(sdk): reuse serial creation header snapshots 2026-09-10 16:18:32 +08:00
_Kerman efa5294c00 test(agent-loop): drop duplicate creation disposal case 2026-09-10 16:18:25 +08:00
_Kerman 11656ca683 refactor(agent): keep registry registration startup-only 2026-09-10 16:18:04 +08:00
07akioni ae7462c526 test: clarify live SDK smoke completion acknowledgement 2026-09-10 16:13:54 +08:00
lsdsjy f63fdd89f4 Merge pull request #3889 from deepseek-harness/turn-duration-hours-unit
feat(web): add hours to turn duration labels
2026-09-10 16:12:36 +08:00
creatixchu e17a736cc3 docs: 同步请求图片投影文档与 Agent Note
更新 attachment、attachment-local、llm-deepseek 的 README,附件与 LLM 流式子系统文档,配置目录,type-equiv 清单,以及双语配对记录;新增请求图片投影的 Agent Note,并修正 v41 计价 Note 里关于 640,000 像素预算的说明。
2026-09-10 16:09:20 +08:00
creatixchu 06c491508f fix(llm-deepseek): 请求图片按 V4.1 token 网格投影
ImageRequestPolicy 改为投影规则加可选单边上限加字节目标,投影是 pixel-budget 或 token-grid 的封闭联合。token 网格求解器放在 dsh-attachment 的 request-projection 里,deepSeekImageTokens 复用同一个求解器计价。DeepSeek 路由默认按 14 px patch、3:1 降采样、1024 token 上限的网格投影,每张请求图片加 4096 像素单边上限,字节目标提高到 2 MiB;显式 imagePixelBudget 与 pi-ai 路由仍走总像素预算。缩放只按长边给定,短边由编码器四舍五入,与投影预测一致。变换版本升到 request-image-v6。

Closes #3929
2026-09-10 16:09:18 +08:00
Dudu-0223 b799912e48 Merge remote-tracking branch 'origin/master' into fix/team-identity-context 2026-09-10 16:04:04 +08:00
Dudu-0223 4491da1374 fix(agent-team): include identity only in the initial task 2026-09-10 16:03:49 +08:00
lsdsjy 9098a58f94 refactor(loader-smoke): model the smoke cwd as a discriminated input
Address the review suggestion that a caller-provided `cwd` left
`tempDirPrefix`/`tempDirParent` as dead but still-required inputs. The options
are now a union of "expand a prefix into a fresh temporary cwd" and "reuse this
existing directory", so the type system rejects passing both; the two callers
that reuse a directory drop their unused prefix.
2026-09-10 16:03:46 +08:00
lsdsjy 8d56babe15 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-10 16:01:12 +08:00
lsdsjy 9bcb1e214c fix(headless): refuse a live Agent identity and finish the adopt-only docs
Address the ds-review-bot findings on 472554a29e:

- resolveAgent refuses an Agent already live under the requested id: its owner
  may still drive it, and `whenIdle` is not a single-message signal, so the
  one-shot runner cannot own an exclusive run interval over it. The adoptability
  rules still run first so the refusal stays specific.
- Refresh the README intro, identity section, and limitations, and the Agent
  Note live-identity rule and change scope, for adopt-only.
- Clarify the loader-smoke `cwd` JSDoc (the prefix/parent inputs are ignored)
  and pin the resumed wake's final text in the two-wake e2e.
2026-09-10 16:00:53 +08:00
creatixchu d0ba1a3c6c Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu 2026-09-10 16:00:13 +08:00
creatixchu 7910cd9e0b test(web): assert the shipped feedback command identity 2026-09-10 16:00:08 +08:00
Turtle ac495c0b5f Merge pull request #3934 from deepseek-harness/turtle/fix-weighted-approval-run-identity
fix(ci): refresh weighted approvals after numbered review runs
2026-09-10 15:58:39 +08:00
turtle1999 e7b8ebedaf fix(ci): identify approval review runs by workflow path 2026-09-10 15:55:00 +08:00
Dudu-0223 6e94ee4fb9 test(workspace): pin persisted mode ordering in Web replay 2026-09-10 15:54:57 +08:00
_Kerman c428e6ffb4 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3828 2026-09-10 15:54:56 +08:00
lsdsjy f0724ba7e8 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-10 15:54:21 +08:00
lsdsjy 249d583892 feat(web): add hours to turn duration labels
Turns running at or past 3600s now read 1小时05分03秒 / 1h 05m 03s in the
running clock, the settled Ran-for footer, and the turn-time dialog, which
share formatRunDuration. Below an hour the second and minute branches are
unchanged, and the hour rolls at the unit boundary so 60分00秒 never appears.
2026-09-10 15:51:43 +08:00
Dudu-0223 8423b270e3 fix(workspace): derive recency independently of manual order 2026-09-10 15:51:42 +08:00
_Kerman c09574fc5f Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3828 2026-09-10 15:51:42 +08:00
07akioni 9f27a45408 test: preserve immediate disposal and observe automatic terminal cleanup 2026-09-10 15:49:57 +08:00
creatixchu aed83d29ba Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu 2026-09-10 15:49:55 +08:00
creatixchu fff1ed7aa2 fix(plan): keep command identity independent of optional runtime 2026-09-10 15:48:25 +08:00
c31ac4fdfe feat(web): gate agent preset selection behind a setting (#3870)
* feat(web): gate agent preset selection behind a setting

* refactor(web): trim preset visibility diff

* refactor(web): simplify agent preset selection policy

* test(web): cover agent preset policy branches

* refactor(web): tighten agent preset policy flow

* fix(web): preserve preset default across picker toggle

* docs(web): align agent preset picker docs with the blank-task sync

Both picker transitions carry the effective default to the current blank task; started and historical sessions keep their labels, compositions, and recorded history. The Host README states the client rule and links the Web switch; the Agent Note English side matches the Chinese at line 64. Three i18n pairs re-recorded; doc-sync 34/34.

* refactor(web): simplify agent preset picker policy

* test(web): cover preset policy recovery

* docs(web): clarify disabled preset default copy

* fix(web): reconcile preset policy races

* fix(web): satisfy preset load lint

* fix(web): serialize preset roster refreshes

* fix(web): release preset refresh before notifying waiters

---------

Co-authored-by: ZiyaZhang <[email protected]>
Co-authored-by: Chinesezjc <[email protected]>
2026-09-10 07:48:06 +00:00
creatixchu 81d017bbaa docs: align generated catalog source links in both locales 2026-09-10 15:47:17 +08:00
creatixchu ca827aa2ad fix(commands): refresh catalogs and use type-only export identity 2026-09-10 15:44:30 +08:00
creatixchu 07413c2cc2 Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu 2026-09-10 15:38:30 +08:00
creatixchu 996278e6ce refactor: separate command identity and composer file action ownership 2026-09-10 15:38:13 +08:00
lsdsjy 349a84e81d Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-10 15:31:53 +08:00
turtle1999 d76704170b refactor(home-paths): simplify cache path dispatch 2026-09-10 15:24:27 +08:00
CreatixChu 646c4c2f33 Merge pull request #3924 from deepseek-harness/worktree/image-token-v41
fix(llm-deepseek): 图片 token 预估器改用官方计算器 v41 配置
2026-09-10 15:22:37 +08:00
lsdsjy 127d2c8eb5 fix(headless): align the adopt-only docs and prove resume end to end
Address the ds-review-bot findings on 9aea4b0804:

- Drop the stale "freshly created" wording from the resolveAgent `@returns`
  and rewrite its leading comment for the two remaining identity paths.
- Rewrite the README run-flow paragraph (EN and ZH), the intro sentence, and
  the cordis.patch.yml header comment for adopt-only.
- Add a real-composition two-wake e2e: capture the generated id from the first
  `--json` wake, resume it by `--session-id`, and assert one persisted log holds
  both tasks. `runLoaderSmoke` gains an optional caller-owned `cwd` so both
  wakes share one world; the harness leaves that directory in place.
2026-09-10 15:19:17 +08:00
turtle1999 3407d15bb3 Merge master into the Cordis revert branch 2026-09-10 15:18:28 +08:00
creatixchu ba2a931a23 Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu 2026-09-10 15:18:02 +08:00
turtle1999 7f005e2c0f test(attachment-local): isolate fallback cache home 2026-09-10 15:17:35 +08:00
07akioni 6becec9502 Merge master and retain its strict Linux startup cancellation handling 2026-09-10 15:13:38 +08:00
turtle1999 d911a7b422 feat(attachment-local): move request images into shared cache 2026-09-10 15:08:46 +08:00
07akioni 6d103f8f7c Revert Windows dialog smoke acceptance of unavailable COM classes 2026-09-10 15:08:38 +08:00
Yichen Jiang 61b02efa84 Merge pull request #3916 from deepseek-harness/worktree/fix-deliverable-sidebar-preview
fix(web): preview delivered file links in the sidebar
2026-09-10 15:06:09 +08:00
07akioni d570f32f32 Merge latest master into fix/bundle-speed 2026-09-10 15:03:23 +08:00
Yichen Jiang 8d1e8acdec Merge pull request #3917 from deepseek-harness/worktree/composer-reference-previews
feat(web): preview file and skill references in the sidebar
2026-09-10 15:00:37 +08:00
Yichen Jiang b79a227cec fix(subprocess): preserve Linux cancellation before bootstrap consumption 2026-09-10 14:45:45 +08:00
lsdsjy ea84ad31b6 feat(headless): require --session-id to name an existing Session
`--session-id` now adopts only: an id with no stored Session fails before the
task runs instead of creating an empty history the caller believes it is
resuming. A first round omits the flag, and the runtime reports the minted
`session-<uuid>` identity in the opening `session` event, so a supervisor
persists it and names it on every later wake.

Updates the help text, the README pair, the Agent Note pair, the generated
config catalog, the unit suite, and the product-profile e2e.
2026-09-10 14:44:44 +08:00
creatixchu f657a3e87a Merge remote-tracking branch 'origin/master' into worktree/image-token-v41 2026-09-10 14:42:07 +08:00
creatixchu f24bc3c832 fix(llm-deepseek): correct image pricing documentation and edge-case evidence 2026-09-10 14:40:50 +08:00
07akioni a62fab00b7 test: distinguish unavailable Windows dialog classes in native smoke 2026-09-10 14:33:20 +08:00
07akioni 4c024cf252 fix: preserve early process cancellation and sample stream markers promptly 2026-09-10 14:28:18 +08:00
turtle1999 85df0e7632 docs(home-paths): align cache helper example 2026-09-10 14:24:44 +08:00
lsdsjy 25f95302fd Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run
# Conflicts:
#	packages/bundle/headless/README.i18n.yaml
#	packages/bundle/headless/README.zh.md
2026-09-10 14:20:35 +08:00
Yichen Jiang 24e63e4555 fix(web): handle pending and linked skill previews 2026-09-10 14:19:01 +08:00
turtle1999 e4e0e78dd2 feat(home-paths): add dshCachePath 2026-09-10 14:16:35 +08:00
07akioni 82293ff45d refactor(desktop): remove unused plugin classification query 2026-09-10 14:07:51 +08:00
07akioni d55b2b3dad fix(desktop): make startup recovery and package rebuilds retryable 2026-09-10 14:05:44 +08:00
creatixchu a64dc3a690 fix(llm-deepseek): 图片 token 预估器改用官方计算器 v41 配置
DeepSeek 图像理解指南改为 544×544 放大下限、约 1300×1300 缩小目标、
单图 1024 token 上限;官方图片 Token 计算器以 v41 配置承载该规则。
重写 image-tokens.ts 为 v41 的逐句移植:取消对齐 pad、宽高比钳制和
奇偶校正网格布局,估算值从最坏情况上界变为精确值。测试向量按官方
计算器重新固定,request-pricing 期望值、README 与 Agent Note 同步更新。
2026-09-10 14:05:04 +08:00
Yichen Jiang 5445ad16eb test(web): update file link gallery preview label 2026-09-10 14:01:01 +08:00
Chinesezjc 40606d1558 Merge pull request #3873 from deepseek-harness/ci/failover-canary-leg
ci: opt-in canary runner leg for the CI pool failover switches
2026-09-10 13:59:53 +08:00
Yichen Jiang 6469b522c1 Merge pull request #3899 from deepseek-harness/worktree/manifest-plugin-metadata
feat(manifest): separate package metadata from DSH declarations
2026-09-10 13:55:51 +08:00
creatixchu 6fe445ea08 Merge origin/master into worktree/composer-plus-menu 2026-09-10 13:53:45 +08:00
07akioni 13fc0a60ec Merge remote-tracking branch 'origin/master' into fix/bundle-speed 2026-09-10 13:53:29 +08:00
creatixchu 1faab27329 Merge origin/master into PR #3440 2026-09-10 13:48:38 +08:00
Yichen Jiang 71046d8306 fix(web): unify inline file preview labels 2026-09-10 13:41:18 +08:00
Yichen Jiang ce7ceb6071 refactor(manifest): remove discovery categories 2026-09-10 13:38:36 +08:00
turtle1999 2376d210bd Preserve nested plugin errors after Cordis revert 2026-09-10 13:36:05 +08:00
Yichen Jiang 4115e8867b test(web): update SVG delivery sidebar snapshot 2026-09-10 13:31:36 +08:00
Yichen Jiang ff153dfb12 Merge remote-tracking branch 'origin/master' into worktree/fix-deliverable-sidebar-preview 2026-09-10 13:29:38 +08:00
turtle1999 a38c302e4a Merge master and resolve revert documentation conflicts 2026-09-10 13:09:19 +08:00
Dudu-0223 d769e2d157 docs(agent-team): regenerate the message dependency graph 2026-09-10 12:50:23 +08:00
Dudu-0223 271268a440 test(agent-team): avoid racing initial admission settlement 2026-09-10 12:46:05 +08:00
Dudu-0223 f0201f73a1 Merge master into fix/team-identity-context 2026-09-10 12:44:29 +08:00
Dudu-0223 b76fe343fe fix(agent-team): keep member identity in durable reminders 2026-09-10 12:42:44 +08:00
Tianyi Cui aa8262ec09 Merge pull request #3920 from deepseek-harness/feat/add-readme-citation
docs: add BibTeX citation to READMEs
2026-09-10 12:34:48 +08:00
07akioni dcec949f33 fix(python): allow workspace-only patches during runtime deploy 2026-09-10 11:50:46 +08:00
Yichen Jiang 597f0f8f1f fix(web): align composer reference heights and baselines 2026-09-10 11:50:41 +08:00
fz 81239767a7 docs: add BibTeX citation to READMEs 2026-09-10 11:49:17 +08:00
07akioni 710d7b5c33 Merge latest fix/bundle-speed and resolve desktop documentation conflicts 2026-09-10 11:37:57 +08:00
Yichen Jiang 081e054c54 Merge remote-tracking branch 'origin/master' into worktree/manifest-plugin-metadata
# Conflicts:
#	packages/boot/app-boot/README.i18n.yaml
#	packages/boot/app-boot/README.zh.md
2026-09-10 11:35:22 +08:00
Yichen Jiang 11d6bd05f3 feat(web): preview file and skill references in the sidebar 2026-09-10 11:31:27 +08:00
07akioni 8ce9b21ea9 Merge master into fix/bundle-speed and preserve desktop runtime changes 2026-09-10 11:25:04 +08:00
_Kerman f570a9d8c6 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3869
# Conflicts:
#	docs/config-catalog.i18n.yaml
#	packages/core/system-prompt/README.i18n.yaml
#	packages/core/system-prompt/README.zh.md
#	packages/core/tools/README.i18n.yaml
2026-09-10 11:19:40 +08:00
_Kerman b511cd6c2d Merge origin/master into xtr/trajectory-json-string-wrapping
Resolve the markdown-preview revert and README proofreading against this
PR's trajectory inspection work:

- CodeBlock keeps `showHeader` (the PTC code inspector supplies its own
  toolbar) on top of master's `contentRef` scrollport wrapper.
- ui-primitives README records both `contentRef` and `showHeader`.
- ui-trajectory README keeps master's proofread wording alongside this
  PR's PTC inspection and timing paragraphs.
- Re-record the README.i18n.yaml pairing hashes for the three merged pairs.
2026-09-10 11:19:39 +08:00
_Kerman eb0a666415 Merge remote-tracking branch 'origin/master' into dshw/pr-deepseek-harness-deepseek-harness-3828
# Conflicts:
#	docs/config-catalog.i18n.yaml
#	docs/subsystems/session.i18n.yaml
#	packages/core/session/README.i18n.yaml
2026-09-10 11:19:25 +08:00
Yichen Jiang ca2a9017ef fix(web): preview delivered file links in the sidebar 2026-09-10 11:13:43 +08:00
07akioni 20374de37d Merge remote-tracking branch 'origin/fix/bundle-speed' into fix/parallel-macos-notarization 2026-09-10 11:03:18 +08:00
07akioni 7814ccdfd9 Merge fix/bundle-speed into fix/parallel-macos-notarization 2026-09-10 11:02:53 +08:00
Yichen Jiang 66e4f809d7 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-10 10:59:10 +08:00
Yichen Jiang 80badab556 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal
# Conflicts:
#	packages/api/README.i18n.yaml
#	packages/api/README.zh.md
#	packages/e2b/subprocess-e2b/README.i18n.yaml
#	packages/e2b/subprocess-e2b/README.zh.md
2026-09-10 10:52:51 +08:00
Tianyi Cui 2377c272a8 Merge pull request #3908 from deepseek-harness/docs/readme-bilingual-proofread-20260909
docs(i18n): proofread bilingual README corpus
2026-09-10 10:50:25 +08:00
Yichen Jiang 12d3be3514 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal
# Conflicts:
#	packages/client/ui-sidebar-right/README.i18n.yaml
#	packages/client/ui-sidebar-right/README.md
#	packages/client/ui-sidebar-right/README.zh.md
2026-09-10 10:48:47 +08:00
j-xiang 4ddf777ad8 Merge master into README proofreading branch 2026-09-10 10:42:13 +08:00
Yichen Jiang 4af56cf808 feat(llm): sync V41 Messages catalog and keep Web on Chat Completions 2026-09-10 10:32:04 +08:00
Yichen Jiang 5c82aa9371 refactor(manifest): separate package metadata from public DSH fields 2026-09-10 10:28:47 +08:00
Yichen Jiang 61176bf7c6 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages
# Conflicts:
#	apps/web/tests/scaffold.ts
2026-09-10 10:25:39 +08:00
imccyu 183f08e9c6 Merge pull request #3912 from deepseek-harness/worktree/release/dsh-0.1.5-rc.1
release: [email protected]
2026-09-10 09:36:48 +08:00
imccyu 1ef9c1fa9a release(dsh): 0.1.5-rc.1 2026-09-10 09:13:14 +08:00
imccyu ecc6f11cf1 Merge pull request #3824 from deepseek-harness/worktree/deepseek-flash-catalog
feat(llm): add V41 Flash alongside V4 models
2026-09-10 09:13:00 +08:00
Yichen Jiang 0729dbec66 feat(llm): restore V4 Flash Vision Exp catalog entry 2026-09-10 03:39:57 +08:00
Yichen Jiang 441385fe38 feat(llm): retain V4 models alongside V41 Flash 2026-09-10 03:39:57 +08:00
Yichen Jiang c98fe6ecdd test(web): pin the recorded model before scenario overlays 2026-09-10 03:39:57 +08:00
Yichen Jiang 719331ecb7 test: keep webhook smoke on the shipped model route 2026-09-10 03:39:56 +08:00
Yichen Jiang 61c5954372 test: pin model routes in webhook and Cordis fixtures 2026-09-10 03:39:56 +08:00
Yichen Jiang bc5fd3b8dc feat(llm): default Chat Completions to DeepSeek V41 Flash 2026-09-10 03:39:56 +08:00
imccyu 4cbaeb8d69 Merge pull request #3909 from deepseek-harness/fix/sidebar-preview-layout-polish
fix(client): refine sidebar previews, guide, and stats
2026-09-10 03:38:01 +08:00
imccyu 3435dbd690 fix(web): address sidebar and stats review feedback 2026-09-10 03:20:55 +08:00
j-xiang e14b8795a9 docs(i18n): fix targeted README review findings 2026-09-10 03:04:19 +08:00
imccyu 57da0906f3 Merge remote-tracking branch 'refs/remotes/origin/pr-3910' into fix/sidebar-preview-layout-polish 2026-09-10 02:04:17 +08:00
j-xiang 267e8b933b docs(i18n): address README review feedback 2026-09-10 02:03:18 +08:00
imccyu ded26f225d chore(client): refresh slot catalog 2026-09-10 01:57:14 +08:00
imccyu 19b3c4c34b fix(client): keep preview scrollports current 2026-09-10 01:51:21 +08:00
j-xiang 89743f33b2 docs(i18n): polish reviewed README terminology 2026-09-10 01:42:58 +08:00
Yif 53d5c9386e fix(sidebar): pin the guide hero to the pure-gray ramp
The label-dimmed alias rides the bluish neutral ramp, whose cast shows
at the 56px compass; the hero now takes neutral-200 (light) and
neutral-700 (dark) from the pure-gray static ramp.
2026-09-10 01:32:45 +08:00
j-xiang b5578ae46a Merge master into README proofreading branch 2026-09-10 01:29:37 +08:00
Yif 1c044f5140 feat(sidebar): refine guide start page and quiet stat pill noise
The guide tab gains a compass hero, per-entry descriptions (shown while
at most four entries are listed), and a quieter cube placeholder for
entries that registered no glyph; the files entry registers its
description and folder glyph. Bundled stat fixes: the session usage
dialog drops the zero cache-write row, and the turn-time pill renders as
plain text when its dialog would only repeat the pill's own duration.
2026-09-10 01:23:24 +08:00
imccyu 284ab15519 fix(client): refine sidebar document previews 2026-09-10 01:04:55 +08:00
j-xiang 8e8fb2fc62 docs(i18n): proofread bilingual README corpus 2026-09-10 00:21:32 +08:00
imccyu b2e3b2a012 Merge pull request #3893 from deepseek-harness/worktree/release/dsh-0.1.5-alpha.2
release: [email protected]
2026-09-09 22:13:03 +08:00
imccyu dd393c1820 release(dsh): 0.1.5-alpha.2 2026-09-09 22:02:09 +08:00
Tianyi Cui 7588e1ea11 Merge pull request #3901 from deepseek-harness/revert-3710-feat/mermaid-preview
Revert "feat(web): preview Mermaid, Graphviz, SVG, and HTML code fences"
2026-09-09 22:01:43 +08:00
Tianyi Cui 58956c1a8a Revert "feat(web): preview Mermaid, Graphviz, SVG, and HTML code fences" 2026-09-09 22:01:10 +08:00
07akioni d09b4accd3 feat: 打包 + 错误处理优化 2026-09-09 21:10:02 +08:00
lsdsjy a69933d3e5 fix(headless): omit step usage when any attempt lacks a sample
Address the ds-review-bot v12 critical on PR #3849: `addUsage` published the
sum of the attempts that reported usage while silently dropping the ones that
did not, so `step_end.usage` looked exact but under-counted billed tokens.
Track completeness per step and omit usage entirely when any attempt in the
step reports no sample, matching `deriveTurnTokenUsage`'s fail-closed rule.
2026-09-09 21:05:19 +08:00
Yichen Jiang 220678e57d feat(manifest): declare plugin format, tags, and host compatibility 2026-09-09 21:03:44 +08:00
lsdsjy 980b4b77af fix(headless): close the v11 review gaps in the json run surface
Address the ds-review-bot v11 review on PR #3849:

- Accumulate every attempt's usage into `step_end.usage`, including a retried
  attempt whose only sample sits in its discarded `assistant/attempt` stream.
- Re-check the adopted Session after the idle wait, so a preset selected in
  that window is still rejected.
- Reject a whitespace-only `sessionId` supplied through a Cordis overlay, which
  bypasses the CLI trim check.
- Narrow the documented error contract: a profile whose own plugins fail to
  load exits before the runner mounts and keeps only stderr diagnostics.
2026-09-09 20:46:16 +08:00
_Kerman 192cc03dc4 Merge origin/master and preserve PTC code inspector controls 2026-09-09 20:35:12 +08:00
Yudong Han 83253e0d16 Merge pull request #3710 from deepseek-harness/feat/mermaid-preview
feat(web): preview Mermaid, Graphviz, SVG, and HTML code fences
2026-09-09 20:24:43 +08:00
_Kerman 9087b275bf Merge remote-tracking branch 'origin/master' into HEAD 2026-09-09 20:22:14 +08:00
imccyu 9d83e3e534 Merge pull request #3890 from deepseek-harness/feat/sidebar-image-preview
feat(sidebar): preview image files
2026-09-09 20:17:43 +08:00
lsdsjy b3f6c5e991 fix(headless): count the line terminator inside the 32 KiB event cap
Address the ds-review-bot v10 review on PR #3849: `boundJsonLine` bounded the
serialized JSON but not the newline the writer appends, so an event that
serialized to exactly 32 KiB produced a 32 KiB + 1 byte record. Reserve the
terminator byte inside the cap, document the cap as newline-inclusive, and
assert the written record length in the boundary tests.
2026-09-09 20:16:58 +08:00
Chinesezjc 360354e38b Merge remote-tracking branch 'origin/master' into ci/failover-canary-leg 2026-09-09 20:04:23 +08:00
_Kerman 5e0e5d00a6 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-09 20:02:08 +08:00
imccyu 8870a13aa9 feat(sidebar): preview image files 2026-09-09 20:01:24 +08:00
Tianyi Cui f05e0fb6b9 Merge pull request #3878 from deepseek-harness/fix/ci-cancel-superseded
fix(ci): cancel superseded PR and master validation
2026-09-09 20:00:13 +08:00
Yudong 7aa73877e6 Merge origin/master into feat/mermaid-preview 2026-09-09 19:53:42 +08:00
Yichen Jiang cfdfd7ccb9 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal 2026-09-09 19:53:05 +08:00
lsdsjy 755b44bbcd fix(headless): require the query service for every --session-id run
Address the ds-review-bot v9 review on PR #3849:

- Require `sessionQuery` before the live fast path too: a later process has no
  live Agent and must find the id through the query service, so a live identity
  no longer bypasses the check.
- Soften the live `stat` comment to what it proves: a stored record rules out an
  in-memory registration, while write-handle ownership is not queryable.
- Complete the `parseArguments` JSDoc and fix the broken English sentence in
  README.md and the Agent Note.
2026-09-09 19:53:03 +08:00
Yichen Jiang d28625c6d5 fix(web): restore terminal snapshots across host and sidebar lifecycles 2026-09-09 19:52:55 +08:00
_Kerman 29c09d320f refactor(client): remove unused JSON disclosure locale keys 2026-09-09 19:52:42 +08:00
_Kerman b6858789f0 fix(trajectory): identify PTC calls from recorded tool metadata 2026-09-09 19:52:42 +08:00
_Kerman 2bc1a7e347 test(trajectory): pin thinking defaults when switching records 2026-09-09 19:52:42 +08:00
_Kerman e7fc4e8fdc fix(client): isolate JSON copy feedback and guard string measurement 2026-09-09 19:52:42 +08:00
_Kerman 573b044320 refactor(client): share the line wrapping icon 2026-09-09 19:52:42 +08:00
Yudong 516bfd5f30 test(web): cover preview import recovery and color schemes 2026-09-09 19:49:56 +08:00
creatixchu 5b8c445eeb fix(snapshots): preserve released inline image fixture 2026-09-09 19:46:13 +08:00
lsdsjy 7a4a55f30b fix(headless): fail closed on malformed preset records and unroundtrippable args
Address the ds-review-bot v6 review on PR #3849:

- A persisted `agent-preset/selected` record whose `data.agentPreset` is not a
  non-empty string used to read as "no preset", letting the run continue under
  this bundle's composition. It now fails closed.
- A live identity now has to be known to `sessionPersistence` (`stat`), since
  an Agent registered only in memory has no write handle and would flush
  nothing.
- Tool arguments that JSON cannot round-trip (an overflowing number such as
  `1e400`) keep their raw text instead of the `null` JSON.stringify reports.
2026-09-09 19:39:31 +08:00
creatixchu 2025036158 Merge origin/master into PR #3440 2026-09-09 19:36:15 +08:00
_Kerman b11c94716f Merge remote-tracking branch 'origin/master' into HEAD 2026-09-09 19:25:21 +08:00
imccyu 16953f2b8e Merge pull request #3884 from deepseek-harness/feat/public-experimental-agent-team
release: publish experimental Agent Teams packages
2026-09-09 19:21:51 +08:00
imccyu fbfc8123e3 Merge pull request #3871 from deepseek-harness/worktree-fixworkerflock
fix(webworker): restore session and workspace writes
2026-09-09 19:21:07 +08:00
Chinesezjc acc8535390 docs(ci): record the canary failover value and the glob fix across internal notes
The failover runbook, the node-compatibility note, the native-Windows note,
and the serial-reference note record the third variable value's scope and
exclusions (bilingual, pairing re-recorded); a new implemented process note
owns the value's participating set, the dormant-by-default canary posture,
and the walker-fix rationale. .github/AGENTS.md points at both notes.
2026-09-09 19:20:29 +08:00
Chinesezjc d79294f23a fix(scripts): enumerate repo files with a dirent walker instead of node fs.glob
Node's internal fs.glob, from some 24.x releases, lstat-probes
<matched>/<next segment> while expanding a ** pattern over a symlinked
file and throws ENOTDIR instead of skipping (observed on node 24.13.0
scanning snapshots/acp/image-compaction's symlinked
system-prompt.expected.md, which failed verify-md-wrap inside
check:ci:static). uniqueRepoFiles now walks with dirent types, never
probing a path under a file, and mirrors fs.glob's dot:false and
follow:false semantics: the four repo-files gate pattern sets enumerate
identically. Broken or cyclic symlinks fail loudly in the caller's
realpathSync, and pattern forms the walker does not model (metacharacters,
a trailing **, empty segments) are rejected up front. repo-files.spec.ts
pins the symlink, broken-link, and rejection behavior.
2026-09-09 19:20:29 +08:00
07akioni fc7abdcfb5 feat: 优化 windows 安装包打包 2026-09-09 19:18:34 +08:00
lsdsjy a5c21a69b8 fix(headless): require persistence for every --session-id path
Address the ds-review-bot v6 review on PR #3849: the guard sat before the
create branch only, so adopting an already-live identity could still run a
session whose history dies at exit. Move it to the top of resolveAgent, where
it covers both the live and the create path, and add a rejection test for the
live path alongside the existing create-path one.
2026-09-09 19:17:01 +08:00
Yichen Jiang f030b544a9 Merge remote-tracking branch 'origin/master' into worktree/web-sidebar-terminal
# Conflicts:
#	packages/client/ui-sidebar-right/README.i18n.yaml
#	packages/client/ui-sidebar-right/README.md
#	packages/client/ui-sidebar-right/README.zh.md
2026-09-09 19:15:51 +08:00
Chinesezjc 9bae613128 ci: add an opt-in canary value to the CI pool failover switches
The per-platform failover repository variables (DSH_CI_FAILOVER_LINUX and
DSH_CI_FAILOVER_WINDOWS) accept a third value alongside the existing
selfhosted routing. Set by a repository writer, the value retargets the
always-on pull-request and master lanes onto the matching hosted-runner
labels (16-core pools map to the 16-vCPU tier, standard legs to 4-vCPU);
unset or any other value keeps today's runners, so the leg is dormant until
a writer opts in - a canary path for capacity fallback, never the default.

Participating jobs: the three enterprise Linux workers, the node-compat
legs, the four Windows lanes, and the all-checks-passed verdict in ci.yml;
expected-filenames; the sandbox bwrap leg; and the two manual benchmark
matrices (4-32 vCPU tiers map to equivalents; 64/96-core rows keep the
fleet labels). Everything pinned to its own runner by spec, credential, or
host capability keeps it under every value, and the release-shaped chains
stay entirely hosted. request-review.yml was retired on master while this
work was in flight and the branch takes the deletion.

Specs pin the new branches: ci-workflow.spec.ts asserts the labels appear
in the extended selectors, and ci-compatible-selfhosted.spec.ts evaluates
the routing mode.
2026-09-09 19:02:10 +08:00
lsdsjy 97e4565a04 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-09 19:01:51 +08:00
lsdsjy b3756a89d1 fix(headless): fail loud when --session-id would create a non-durable session
Address the ds-review-bot v6 review on PR #3849:

- Require the composed sessionPersistence service before creating an exact
  --session-id identity, so a profile that keeps sessionQuery but drops
  persistence exits 1 instead of printing an id whose history dies at exit.
- Stop exporting the now-internal boundJsonEvent and describe it in the
  present tense; its test now exercises boundJsonLine.
- Say "64 levels or deeper" in the README and Agent Note, matching the
  depth budget (an input keeps 63 nested containers, the 64th is cut).

Tests: 77 headless unit tests with per-file 100% coverage on src, 14 profile
e2e tests, constraints/typecheck/lint/doc/pairing gates.
2026-09-09 19:00:06 +08:00
_Kerman af20a0f004 Merge remote-tracking branch 'origin/master' into HEAD
# Conflicts:
#	packages/client/ui-primitives/src/markdown/CodeBlock.tsx
2026-09-09 18:57:48 +08:00
imccyu 63187de80f release: publish experimental Agent Teams packages 2026-09-09 18:54:14 +08:00
imccyu 9b308c9568 Merge pull request #3819 from deepseek-harness/feat/artifact-file-actions
feat(web): unify file icons, delivery actions, and sidebar interactions
2026-09-09 18:51:34 +08:00
_Kerman a9b5faef0b test(chat): collapse live thinking before checking its tail 2026-09-09 18:49:03 +08:00
_Kerman 4da3102130 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-09 18:48:34 +08:00
Ruilin Geng 2b521b31f1 Merge pull request #3846 from deepseek-harness/fix/mcp-pagination-cycles
fix(mcp): reject repeated tool discovery cursors
2026-09-09 18:46:44 +08:00
_Kerman 508490b531 test(trajectory): omit absent optional fixture fields 2026-09-09 18:35:07 +08:00
lsdsjy a6fa383ba8 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-09 18:34:16 +08:00
_Kerman 9c6b282535 test(client): cover raw JSON strings and wrapping preferences 2026-09-09 18:34:12 +08:00
turtle1999 d225dbba50 Remove nonessential remnants of Cordis PR 932 2026-09-09 18:34:11 +08:00
_Kerman 5230c60525 test(trajectory): expect ordinal step detail titles 2026-09-09 18:34:11 +08:00
_Kerman a73aca1978 test(client): cover assistant thinking disclosure defaults 2026-09-09 18:34:11 +08:00
lsdsjy d2dc505830 ci: re-trigger the pull-request run after the superseded run was dropped 2026-09-09 18:28:34 +08:00
_Kerman 056ca664c8 Merge remote-tracking branch 'origin/master' into HEAD 2026-09-09 18:24:55 +08:00
creatixchu 0b3248e67d test(web): stabilize delivered-file snapshots across CI hosts 2026-09-09 18:21:45 +08:00
Yichen Jiang 6e945def2f Merge origin/master into worktree/web-sidebar-terminal 2026-09-09 18:21:21 +08:00
imccyu d7ecdcf02c fix(lint): parenthesize file handle callback 2026-09-09 18:21:17 +08:00
imccyu 60a3d0fbc2 fix(webworker): support file handle chmod 2026-09-09 18:21:17 +08:00
imccyu 950474605f docs(webworker): record file handle compatibility 2026-09-09 18:21:16 +08:00
imccyu a7bd71e024 fix(webworker): preserve bigint file handle identity 2026-09-09 18:21:16 +08:00
imccyu 0a9c774617 docs(webworker): document bigint handle identity 2026-09-09 18:21:16 +08:00
_Kerman f36e1a3d1d fix(trajectory): show turn and ordinal step in detail titles 2026-09-09 18:20:13 +08:00
Tianyi Cui 3c4c6c3d7b fix(ci): cancel superseded PR and master validation 2026-09-09 18:20:01 +08:00
_Kerman f0810edd7f fix(trajectory): expand preview thinking by default 2026-09-09 18:19:51 +08:00
Yichen Jiang 0a525601a7 fix(web): harden sidebar terminal lifecycle and recovery 2026-09-09 18:17:47 +08:00
gengruilin 556857bcd2 Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 18:15:36 +08:00
_Kerman 632b8e19b4 feat(trajectory): add a PTC code inspector 2026-09-09 18:15:09 +08:00
creatixchu b53fed39d7 Merge remote-tracking branch 'origin/master' into worktree/pr-3819-release-sync 2026-09-09 18:07:26 +08:00
creatixchu b9587ada12 fix(web): open delivered files using the viewed session scope 2026-09-09 18:07:16 +08:00
lsdsjy 3488db77bf fix(headless): cap the bounding recursion depth and route error events through the line cap
Address the ds-review-bot v6 review on PR #3849:

- Cut a projected payload nested deeper than 64 levels instead of recursing
  until the stack overflows; the isolated session listener would otherwise
  drop the `tool_call` silently.
- Write both `error` events (startup usage errors and runner failures) through
  `boundJsonLine`, so the documented 32 KiB line cap covers JSON-escaping
  expansion as well as the 8 KiB message cap.

Tests: 76 headless unit tests with per-file 100% coverage on src, 14 profile
e2e tests, constraints/typecheck/lint/doc/pairing gates.
2026-09-09 18:06:29 +08:00
Tianyi Cui 8803287f71 Merge pull request #3853 from deepseek-harness/fix/3835-tool-guidance-restrictions
fix: respect scoped tools in filesystem and web guidance
2026-09-09 18:04:11 +08:00
yudshj 4e0a405a9b fix(web): size diagram previews to image content 2026-09-09 18:01:36 +08:00
creatixchu 2ab73152ae Merge remote-tracking branch 'origin/master' into worktree/pr-3819-release-sync 2026-09-09 18:01:07 +08:00
creatixchu 7f69a3ec37 fix(web): complete shared file icon consumers and browser checks 2026-09-09 18:01:04 +08:00
creatixchu 43bf1d3f93 Merge remote-tracking branch 'origin/master' into worktree/pr-3819-release-sync 2026-09-09 17:56:14 +08:00
creatixchu 454caf0280 fix(web): reject empty sidebar splits and clarify page targeting 2026-09-09 17:56:11 +08:00
imccyu bd60582403 Merge pull request #3859 from deepseek-harness/worktree-sidebarsubagent
fix(workspace-files): support subagent session roots
2026-09-09 17:55:17 +08:00
creatixchu 9961eedec8 fix(web): preserve office file icons and lazy sidebar replay 2026-09-09 17:48:47 +08:00
Tianyi Cui cd95dda1bd Merge pull request #3866 from deepseek-harness/docs/session-format-authority
docs(session): centralize format status and timeless guidance
2026-09-09 17:46:17 +08:00
lsdsjy 0ca42f3b78 fix(headless): bound whole events, re-check adoption after resume, tighten --json scan
Address the ds-review-bot v5/v6 review on PR #3849:

- Register the shared `lib/json-stream-*.js` chunk in `packageFileExtras` and
  order `files` as the workspace-constraints gate requires, so
  `pnpm run constraints` passes.
- Re-validate the preset on the Session `agents.resume` actually attached,
  closing the TOCTOU window between `observeSession` and the write lease.
- Bound one serialized event line at 32 KiB: an over-long event keeps its
  scalar fields, drops structured ones, and at the extreme reduces to `type`.
- Normalize an empty tool-argument string to `{}`, matching the executor.
- Scan only real `--json` flags (stop at `--`, skip a `--session-id` value)
  before installing the JSON error override.

Tests: 74 headless unit tests with per-file 100% coverage on src, 14 profile
e2e tests, constraints/typecheck/lint/doc/pairing gates, tarball import smoke.
2026-09-09 17:40:47 +08:00
HYD@OVERTON 9e050260e4 feat(web): adapt diagram previews to application theme 2026-09-09 17:39:04 +08:00
yudshj fb84576ea7 refactor(web): unify markdown preview components and notes 2026-09-09 17:39:04 +08:00
yudshj a8f9d81f3a feat(client): preview Graphviz, SVG, and HTML Markdown fences 2026-09-09 17:39:04 +08:00
fz 4fabbd39c5 test(web): scope Mermaid snapshots to diagram content 2026-09-09 17:39:04 +08:00
fz 0bfd594875 refactor(ui): use tagged Mermaid preview results 2026-09-09 17:39:04 +08:00
fz 220103670a fix(ui): retain Mermaid previews and support hybrid touch controls 2026-09-09 17:39:04 +08:00
fz fc0c885287 style(ui): simplify Mermaid preview controls 2026-09-09 17:39:04 +08:00
fz ee35e40bc8 feat(ui): preview Mermaid diagrams in Chat code blocks 2026-09-09 17:39:04 +08:00
_Kerman 3a32c617a3 feat(trajectory): remember JSON string wrapping preference 2026-09-09 17:38:02 +08:00
creatixchu 7c9b29563d Merge branch 'worktree/pr-3872-release-sync' into worktree/pr-3819-release-sync 2026-09-09 17:34:20 +08:00
creatixchu c6b3900d2a Merge remote-tracking branch 'origin/master' into worktree/pr-3872-release-sync 2026-09-09 17:33:47 +08:00
turtle1999 2abb542a22 Adapt current consumers to nontransactional Cordis Loader
Move exact user-patch watching into app boot and audit activation explicitly.
Wait for CLI fallback HMR and preset subtrees. Join the first fiber-disposal
result in the directory chooser and browser package runner. Keep async
update errors observable without restoring generic Loader rollback.

Restore the current workspace-link and Schemastery manifests and their gate.
Preserve unconditional patch-input cloning, awaited Include initialization,
and durable teardown writes. Update diagnostics, snapshots, bilingual docs,
and the exhaustive vendor divergence ledger.

Validation: focused Loader, watcher, preset, inventory, and cleanup tests;
100% watcher coverage; 32 Web preset cases; real CLI webhook model session;
SDK smokes; 13 headless expected cases; 3 recorded-session replays; build,
typecheck, lint contracts, and all 34 doc-sync gates.

A full Node 24/Python 3.12 Vitest run had 21806 passes and two failures:
a baseline-reproduced experimental CSS import failure and a Python probe
timeout that passed alone. The default Node 26/Python 3.9 run additionally
exposed unsupported Python and FileHandle-GC host failures.
2026-09-09 17:33:32 +08:00
turtle1999 e07f41d5fd Revert #932 transactional Cordis reload changes
Apply real git revert --no-commit operations in reverse order:

acb0797a1421eab128a4bd03355e930c1a221fb0
 af97714f2413a3e55f0f15ab1b283beb133b3f2d
2018b0b7e64b444bff82b62c9466f13adc4603bb
43170c027cfbf9155dad7a43c01ca39364d4aef2
a621e927d6b3763798c3929e28912d86c9576f2d

Resolve the app-boot package move, preserve deleted files and frozen archived
notes, retain the modern lockfile graph, and preserve independent later
lifecycle and lazy-config changes. Do not revert 6ac7d72d: that merge also
contains eight unrelated commits from the #936 dependency chain.

This is the mechanical intermediate result. Its focused check exposed
consumer and packaging incompatibilities. The following compatibility
commit supplies those adaptations and reconciles the vendor ledger.
2026-09-09 17:32:09 +08:00
_Kerman 2f7a02022a Merge origin/master into xtr/agent-setup-hook 2026-09-09 17:31:17 +08:00
imccyu d7dcc5f5af test(web): update sidebar preview expectations 2026-09-09 17:29:38 +08:00
imccyu 177eacbe8b docs(workspace-files): clarify session scope behavior 2026-09-09 17:29:38 +08:00
imccyu c5432b94f9 chore: satisfy workspace files CI gates 2026-09-09 17:29:38 +08:00
imccyu 7214a97723 fix(workspace-files): resolve subagent session roots 2026-09-09 17:29:38 +08:00
_Kerman e1f0911618 test(ptc): strengthen literal prompt assertions 2026-09-09 17:29:01 +08:00
Yichen Jiang 00b16cb63a fix(web): align terminal preview, styles, and snapshots 2026-09-09 17:29:00 +08:00
_Kerman 5a673b8a59 refactor(ptc): share the prompt section return type 2026-09-09 17:28:47 +08:00
Yif 3b4122ac16 feat(client): polish sidebar and file preview interactions
Seed sidebar default pages lazily on expansion, scope page-tab dedupe
to the target pane, let a sole tab split its own pane with a factory
backfill, close menus when focus enters a sandboxed iframe, and pin the
code preview copy banner while unifying its background with the pane.
2026-09-09 17:27:37 +08:00
creatixchu 1d55cab9e6 fix(web): use shared file type icons in delivery cards 2026-09-09 17:23:07 +08:00
creatixchu 644f6abd7c Merge branch 'worktree/pr-3858-release-sync' into worktree/pr-3819-release-sync
# Conflicts:
#	packages/client/ui-primitives/README.i18n.yaml
#	packages/client/ui-primitives/README.md
#	packages/client/ui-primitives/README.zh.md
2026-09-09 17:20:37 +08:00
creatixchu a7b5c8a043 Merge remote-tracking branch 'origin/master' into worktree/pr-3858-release-sync 2026-09-09 17:20:01 +08:00
_Kerman c6fb28a164 fix(client): use a collapse icon for expanded JSON strings 2026-09-09 17:19:18 +08:00
_Kerman 1549483397 feat(client): expand reasoning-only assistant output by default 2026-09-09 17:19:10 +08:00
_Kerman e4f0871b80 fix(trajectory): clarify overview overflow and tighten spacing 2026-09-09 17:19:03 +08:00
_Kerman ac09b7b519 fix(trajectory): narrow Chinese role column 2026-09-09 17:16:15 +08:00
lsdsjy 3797127eb2 fix(headless): publish the shared chunk and read the session's current preset
Address the ds-review-bot v6 review on PR #3849:

- Publish the `lib/json-stream-*.js` chunk that both entry points import.
  Without it the installed tarball failed to load with ERR_MODULE_NOT_FOUND,
  because `files` listed only `lib/index.js` and `lib/startup.js`.
- Resolve a Session's preset from its log (`agent-preset/selected`) instead of
  only the immutable creation header, so a Session that switched preset while
  blank is rejected too. Both the persisted and the live path scan the events.
- Write the `--json` error event for Commander grammar rejections (an unknown
  option, a missing option value) as well as the action's own usage errors, and
  drop Commander's `error: ` prefix from the event message so one event type
  carries one message shape.
- Reject a whitespace-only positional instead of falling back to stdin, so an
  accidental blank argument cannot consume a pipe.
- Document the stdin task as verbatim and the widened usage-error contract.

Tests: 68 headless unit tests with per-file 100% coverage on src, 14 profile
e2e tests, plus a packed-tarball import smoke check.
2026-09-09 17:16:09 +08:00
_Kerman aeb13413c7 test(agent): complete awaited creation migration 2026-09-09 17:14:53 +08:00
_Kerman 0fb509f544 fix(agent): preserve initialization failure during teardown 2026-09-09 17:14:16 +08:00
_Kerman e4be27d1d9 test(ptc): align workspace snapshot with shared SDK prompt 2026-09-09 17:13:08 +08:00
creatixchu f5804826a8 Merge remote-tracking branch 'origin/master' into codex/pr-3819 2026-09-09 17:13:08 +08:00
Tianyi Cui 8a41f7a8e1 docs(session): clarify release authority review findings 2026-09-09 17:12:06 +08:00
_Kerman e779831f40 fix(client): restore assistant timing from recorded streams 2026-09-09 17:10:44 +08:00
Tianyi Cui 56c4c3e47c docs(session): centralize format status and timeless guidance 2026-09-09 17:06:03 +08:00
Tianyi Cui bafe887d6d Merge pull request #3851 from deepseek-harness/docs/released-v3-wording
docs(session): correct released V3 compatibility guidance
2026-09-09 17:05:14 +08:00
creatixchu dfda1d6aaf fix(web): ship protocol runtime dependency for delivered file actions 2026-09-09 17:02:24 +08:00
lsdsjy 45032ea1bd fix(headless): address re-review findings on the machine-readable run surface
- Reject adopting a Session recorded under an agent preset, and one that
  recorded no cwd; align the subagent/forked diagnostic wording.
- Write the bounded JSON `error` event for a --json usage error before exit,
  since the runner never mounts to write it.
- Keep the caller's exact --session-id instead of the trimmed value.
- Bound object keys and copy a literal `__proto__` key as data rather than
  through the inherited setter.
- Reject a lone `-` mixed with other task words.
- Document the in-turn failure contract and step-commit text/thinking timing.
2026-09-09 16:58:24 +08:00
gengruilin 86c61e24e9 Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 16:57:25 +08:00
_Kerman 96a1749db6 fix(ptc): preserve literal braces in generated SDK prompts 2026-09-09 16:52:42 +08:00
creatixchu 99bffa7f9a fix(web): update delivery card snapshot and remove unused button segments 2026-09-09 16:52:04 +08:00
creatixchu b443c4d246 fix(present): resolve file services in live agent scope and refresh snapshots 2026-09-09 16:48:39 +08:00
_Kerman 819c2b2442 feat(client): show expanded JSON strings as raw text 2026-09-09 16:47:16 +08:00
creatixchu 7e4d9edc13 Merge remote-tracking branch 'origin/master' into codex/pr-3819 2026-09-09 16:45:05 +08:00
creatixchu 00bfdfbce6 fix(present): follow session filesystem access for deliverables 2026-09-09 16:44:51 +08:00
yixiangihsiang 1eee75b424 fix(client): satisfy icon validation gates 2026-09-09 16:41:15 +08:00
yixiangihsiang 4ee9e055d5 feat(client): add shared file type icons 2026-09-09 16:41:15 +08:00
Yichen Jiang e15a9b1bec feat(web): add interactive sidebar terminals 2026-09-09 16:40:58 +08:00
_Kerman 0e468a62f6 fix(trajectory): hide timing source in overview 2026-09-09 16:39:16 +08:00
Dudu-0223 291b890a1a fix(agent-team): append durable identity reminders after fork history 2026-09-09 16:32:32 +08:00
Turtle 6d99a8a53a Merge pull request #3794 from deepseek-harness/turtle/issue-3761-plugin-instance-label
fix(web): distinguish same-package plugin instances
2026-09-09 16:29:54 +08:00
_Kerman 18739de7c6 feat(client): improve trajectory JSON string display 2026-09-09 16:28:17 +08:00
Tianyi Cui a5fc275a56 docs(session): address scoped review clarifications 2026-09-09 16:26:49 +08:00
Turtle d6a699be19 Merge pull request #3845 from deepseek-harness/turtle/keep-blocked-approval-pending
Keep blocked weighted approvals pending
2026-09-09 16:20:11 +08:00
_Kerman 9b7a8ccc9f feat(agent): await initialization through agent/created 2026-09-09 16:20:00 +08:00
creatixchu 74bf7aaa2f fix(web): preserve artifact actions across reconnects and keyboard use 2026-09-09 16:18:24 +08:00
fz aed212657a docs: clarify scoped guidance and file observation policy 2026-09-09 16:13:09 +08:00
creatixchu a43fa81f82 test: respect startup and teardown budgets in Windows fixtures 2026-09-09 16:08:19 +08:00
Turtle e9f1725b11 docs: record pending weighted approval blockers 2026-09-09 15:58:21 +08:00
turtle1999 0d44848ba6 Rename weighted approval publisher job 2026-09-09 15:58:21 +08:00
turtle1999 8eede0f60d Keep blocked approvals pending 2026-09-09 15:58:21 +08:00
lsdsjy 94d3273c19 Merge remote-tracking branch 'origin/master' into turtle/issue-3848-headless-machine-readable-run 2026-09-09 15:54:26 +08:00
yx.zhang 057b4879cf feat(web): align presented file cards with design 2026-09-09 15:53:44 +08:00
winewill 9c14f2785f perf(desktop): parallelize macOS notarization 2026-09-09 15:52:46 +08:00
lsdsjy ba6e19809d fix(headless): project committed content and scope live adoption 2026-09-09 15:51:23 +08:00
fz 4029451fa7 test: refresh guidance for the fetch-only composition 2026-09-09 15:50:06 +08:00
fz 9d6c964985 fix: respect scoped tools in filesystem and web guidance 2026-09-09 15:45:54 +08:00
Turtle 2e2489b888 fix(web): trim composition marker from plugin subtitles 2026-09-09 15:41:02 +08:00
Tianyi Cui 7b6b8a2baa docs(session): correct released V3 compatibility guidance 2026-09-09 15:32:55 +08:00
creatixchu a8faac2850 Merge origin/master feedback dialog changes into worktree/composer-plus-menu 2026-09-09 15:31:33 +08:00
lsdsjy 8662afb530 docs(headless): regenerate module graph for the run projection 2026-09-09 15:31:24 +08:00
Turtle 2b631c0646 test(web): update plugin settings golden order 2026-09-09 15:27:18 +08:00
yudshj c83f02ad81 fix(web): defer desktop metadata until delivery cards appear 2026-09-09 15:26:34 +08:00
yudshj 5b671d7df3 fix(web): show artifact paths relative to the workspace 2026-09-09 15:26:34 +08:00
yudshj b27d861741 docs: refresh native file action catalog references 2026-09-09 15:26:34 +08:00
yudshj 4158a96e14 feat(web): add native file actions to artifact cards 2026-09-09 15:26:28 +08:00
creatixchu 5a878f7c3e test(web): await restored theme before changing preference 2026-09-09 15:24:56 +08:00
imccyu da042b78ff Merge pull request #3765 from deepseek-harness/worktree/issue-3515-feedback-dialog
反馈弹窗:点踩与 /feedback 共用一个带分类的弹窗,提交后出 toast
2026-09-09 15:24:24 +08:00
lsdsjy 0e7361a489 test(headless): close per-file coverage on the run projection 2026-09-09 15:19:52 +08:00
lsdsjy d5fe1fb887 docs(headless): regenerate event producer graph and restore model-selection note 2026-09-09 15:17:01 +08:00
lsdsjy ba6a90d2f2 feat(headless): add stdin task, --session-id, and --json run events 2026-09-09 15:12:32 +08:00
creatixchu 883e7f80fc Merge origin/master into worktree/composer-plus-menu 2026-09-09 15:02:42 +08:00
Turtle 1349af8e97 fix(web): give plugin identities a full row 2026-09-09 14:57:45 +08:00
creatixchu 0b5fca5c68 Merge remote-tracking branch 'origin/master' into worktree/issue-3515-feedback-dialog
# Conflicts:
#	apps/web/tests/message-feedback-layout.e2e.ts
2026-09-09 14:56:31 +08:00
Yudong Han ba5ce5700c Merge pull request #3832 from deepseek-harness/fix/file-delivery-present
fix(deliverables): distinguish file changes and require explicit delivery
2026-09-09 14:53:17 +08:00
_Kerman 12ba944f54 test(lint): cover example history reader exemptions 2026-09-09 14:52:15 +08:00
_Kerman 14da43d9ff docs(session): link deprecated readers to the history policy 2026-09-09 14:51:50 +08:00
gengruilin 96ac11885e Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 14:40:06 +08:00
Turtle f42fddad34 Merge pull request #3840 from deepseek-harness/turtle/remove-weighted-approval-cyborg-text
Remove cyborg wording from weighted approval
2026-09-09 14:35:50 +08:00
Tianyi Cui 456162410b Merge pull request #3780 from deepseek-harness/fix/master-ci-3661
test(ci): 固定异步完成观察并收紧导入豁免
2026-09-09 14:35:17 +08:00
CreatixChu 4f0adbc99a Merge pull request #3839 from deepseek-harness/fix/3838-composer-placeholder
fix(web): 输入空白字符后隐藏 Composer 占位提示
2026-09-09 14:34:43 +08:00
Huanqi Cao cc5a7697e7 Merge pull request #3698 from deepseek-harness/feat/picker-alt-foreground
fix(picker): foreground-activate the Windows native folder dialog via a synthesized Alt press
2026-09-09 14:28:23 +08:00
turtle1999 1eaba2809b Remove cyborg wording from weighted approval 2026-09-09 14:28:10 +08:00
creatixchu eff482c46b fix(web): stabilize composer command editing and scrolling 2026-09-09 14:27:51 +08:00
creatixchu 6c4a2fb294 Merge remote-tracking branch 'origin/master' into worktree/issue-3515-feedback-dialog
# Conflicts:
#	packages/client/ui-model-selection/tests/browser-plugin.client.spec.ts
2026-09-09 14:27:15 +08:00
creatixchu 224f75dc63 test(web): strengthen composer placeholder regression evidence 2026-09-09 14:15:25 +08:00
yx.zhang 30ea609363 Merge remote-tracking branch 'origin/pr-3765-head' into pr-3765 2026-09-09 14:06:27 +08:00
creatixchu a6ee97ebe1 Merge remote-tracking branch 'origin/master' into fix/3838-composer-placeholder 2026-09-09 14:04:23 +08:00
_Kerman 23f345e1dd Merge remote-tracking branch 'origin/master' into xtr/deprecate-session-event-readers 2026-09-09 14:03:26 +08:00
creatixchu 0f1164cdee fix(web): hide composer placeholder for whitespace drafts 2026-09-09 13:56:54 +08:00
yx.zhang 4b3cfcfaf3 style(feedback): refine feedback dialog disclosure and layout 2026-09-09 13:56:53 +08:00
Turtle cad50c2504 Merge pull request #3792 from deepseek-harness/turtle/issue-3762-provider-base-url-validation
fix(web): validate custom provider base URLs
2026-09-09 13:54:48 +08:00
yudshj 9708c4bb42 test(web): scope SVG delivery snapshot to the chat transcript 2026-09-09 13:41:56 +08:00
yudshj 152e0f42ca fix(web): await settled browser state and shorten file-change label 2026-09-09 13:41:56 +08:00
yudshj 7a0afea27f fix(deliverables): distinguish file changes and require explicit delivery 2026-09-09 13:41:56 +08:00
_Kerman ba1e73ddf5 Revert "test(credentials): await reload commit notifications"
This reverts commit 436ec72d4b744cbf3c1473680388d802940b6b9e.
2026-09-09 13:40:51 +08:00
_Kerman 5367541916 Revert "test(agent-team): await recovered mail acknowledgement"
This reverts commit 6e9b6abf38694f9e2ba4214e87356d90293f8dc8.
2026-09-09 13:40:51 +08:00
Turtle fac33d7ba7 Merge pull request #3791 from deepseek-harness/turtle/issue-3760-zh-display-model-copy
fix(web): localize Chinese display and model copy
2026-09-09 13:36:09 +08:00
pku-xht 6adc878785 Merge master sidebar UI polish 2026-09-09 13:27:07 +08:00
gengruilin 8053f3d450 Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 13:19:42 +08:00
Tianyi Cui 6bc836c0c0 docs(ci): distinguish fixture controls from completion policy 2026-09-09 13:18:09 +08:00
Tianyi Cui 1b56cf6013 test(ci): retain useful regression controls after master rebase 2026-09-09 13:18:09 +08:00
Tianyi Cui da1df501b8 test(runtime): isolate busy-meter decisions from worker startup 2026-09-09 13:18:09 +08:00
Tianyi Cui 43d00069a9 test(web): await restoration before workspace actions 2026-09-09 13:18:09 +08:00
Tianyi Cui bc55f10a46 test(web): wait for admitted queue before steering shortcut 2026-09-09 13:18:09 +08:00
Tianyi Cui d85bdd80c1 test(ci): observe command readiness and transport settlement 2026-09-09 13:18:08 +08:00
Tianyi Cui c93bfee16e test(web): drain retained request handlers before teardown 2026-09-09 13:18:08 +08:00
Tianyi Cui e7feedab5e test(web): await durable UI and responsive layout publication 2026-09-09 13:18:08 +08:00
Tianyi Cui d38391d629 test(web): release timed-out barriers and await frame animations 2026-09-09 13:18:08 +08:00
Tianyi Cui 94a2b0a248 test(ci): await details transitions and report benchmark runtime 2026-09-09 13:18:08 +08:00
Tianyi Cui 5b0d65fd72 test(ci): observe asynchronous completion and settled layout 2026-09-09 13:18:08 +08:00
_Kerman 3744740b62 test(agent-team): await recovered mail acknowledgement 2026-09-09 13:17:21 +08:00
Huanqi Cao 2059bdf78e Merge remote-tracking branch 'origin/master' into feat/picker-alt-foreground 2026-09-09 13:17:20 +08:00
_Kerman c433ac7776 test(credentials): await reload commit notifications 2026-09-09 13:17:20 +08:00
_Kerman a61b0c5a56 test(lint): update the scoped deprecation fingerprint 2026-09-09 13:16:09 +08:00
Turtle ebd7e82075 Merge pull request #3834 from deepseek-harness/turtle/remove-automatic-review-requests
ci: remove automatic review requests
2026-09-09 13:13:25 +08:00
Turtle 939222a4ad fix(web): normalize custom provider base URLs 2026-09-09 13:12:43 +08:00
creatixchu a4954e1bd2 test(web): follow the renamed launcher and capitalized rows in CI gates
Registers slash/input-pick-files with the cordis catalog walk exemptions
beside its sibling slash-input events, and points the web e2e at the
'Add files or run commands' launcher and the capitalized row titles the
paperclip removal and the menu presentation introduced.
2026-09-09 12:58:11 +08:00
Turtle d6ed693f23 ci: remove automatic review requests 2026-09-09 12:56:06 +08:00
_Kerman 8f986c8da6 chore(session): scope history reader exemptions to tests 2026-09-09 12:55:32 +08:00
creatixchu ba066ab5f8 Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu 2026-09-09 12:53:09 +08:00
creatixchu 00c2c47e19 fix(web): key the claim hint on the command name, not the claim token
A localized claim token (/目标 ) made the composer look up hint.目标, miss the
dictionary, and fall back to the Host's usage string. CommandClaim now carries
the catalog name beside the display token, the state projection keeps it, and
the hint lookup reads it.
2026-09-09 12:52:41 +08:00
imccyu 88dfd72537 Merge pull request #3807 from deepseek-harness/sidebar-ui-polish-main
feat(web): right sidebar UI polish
2026-09-09 12:49:57 +08:00
pku-xht 01789bfa4d Merge master benchmark calibration fixes 2026-09-09 12:48:17 +08:00
creatixchu 0199633269 Merge remote-tracking branch 'origin/master' into worktree/issue-3515-feedback-dialog 2026-09-09 12:45:32 +08:00
creatixchu 700e8df91a test(feedback): cover the seed-failure and unmount paths of the Dislike control 2026-09-09 12:45:31 +08:00
_Kerman 37abcb74c5 chore(session): permit existing deprecated history reads 2026-09-09 12:33:53 +08:00
_Kerman aa491acc29 docs(session): record synchronous history read deprecation policy 2026-09-09 12:20:38 +08:00
imccyu 2f2831f033 test(web): align sidebar interaction expectations 2026-09-09 12:15:00 +08:00
imccyu ff3ccb4d7c test(web): use supported sidebar visibility assertion 2026-09-09 12:15:00 +08:00
imccyu 3481d4043e fix(client): align sidebar checks after rebase 2026-09-09 12:15:00 +08:00
Yif d0a1ab8723 chore(client): regenerate slot catalog after master merge 2026-09-09 12:15:00 +08:00
Yif c73dcecb96 fix(textpreview): say a stat-and-read failure once
A file whose stat and first page both fail — outside the workspace, gone —
showed the same line twice: the metadata bar above an identical centred body
failure. The bar now stands only over pages already loaded, and the centred
retry runs the full reload so one press clears both halves.
2026-09-09 12:15:00 +08:00
Yif 284758cbc3 test(web): refresh goldens for the sidebar aria renames
The expand/collapse and wrap toggles' accessible names diverged from their
tooltips, and the replay harness now normalizes clientTimeZone in recorded
sessions; every golden holding the old names or the literal zone refreshes.
2026-09-09 12:15:00 +08:00
Yif c76e371874 fix(sidebar): address review feedback on aria names, touch hits, and fullscreen close
- Split toggle aria-labels from their tooltips: the sidebar expand/collapse
  buttons announce "Open/Collapse right sidebar" so the two sidebars stay
  distinguishable, and the wrap toggle keeps an invariant "Line wrap" name
  beside aria-pressed while the tooltip keeps the action wording.
- A hidden tab close also takes no pointer events, so a touch press on the
  chip's trailing edge activates the tab instead of closing it; e2e closes
  hover the chip first, the way a reader would.
- Closing the last docked tab in fullscreen hands the mode back to push, so
  the next expand does not give the whole window to the start page.
- The drop scrim mixes from bg-base, not bg-layer-2, which dark mode binds
  to a lighter bluish step than the column beneath it.
- Document the menuitem probe in TabMenuExtras and the strip fade's known
  gap; match the open-in-app split-button comment to the 28px control row.
2026-09-09 12:14:59 +08:00
Yif 814f447b8f fix(dockkit): move ui-primitives to devDependencies and hoist zone glyph paths past the i18n gate 2026-09-09 12:14:59 +08:00
Yif 5e44fc79d3 test(web): refresh header and file-preview goldens for renamed controls 2026-09-09 12:14:59 +08:00
Yif 7a032b6366 test(web): disambiguate the left sidebar toggle from the right expand control 2026-09-09 12:14:58 +08:00
Yif 237618ca33 test(dockkit): cover the bottom drop-zone glyph 2026-09-09 12:14:58 +08:00
Yif e94241f601 feat(web): move the session-log control into the header more-actions menu 2026-09-09 12:14:58 +08:00
Yif 5ee343f735 feat(sidebar): live tab titles with file-type icons and disambiguated paths 2026-09-09 12:14:58 +08:00
Yif f4c1e31ba8 feat(sidebar): last-tab close rules and collapsed expand button
The lone guide tab withholds its close and menu, any other lone tab closes
together with the column, and the expand button renders only while collapsed
so the corner seat costs the header nothing.
2026-09-09 12:14:58 +08:00
Yif 838dbbe0d0 feat(dockkit): refine strip and chip presentation, close policy, and room reading
Chip capsules with scroll fades and live titles, drop-zone hint cards, the
canCloseTab control policy with the quiet lone chip and self-dismissing empty
menu, and a room reading that stays independent of hidden split controls.
2026-09-09 12:14:58 +08:00
Yif 69a13cef6d feat(primitives): add file-type icon set and shared link icon 2026-09-09 12:14:57 +08:00
_Kerman 5cfc765ff6 docs(session): deprecate direct event readers 2026-09-09 12:11:13 +08:00
Turtle bc4fde06ac fix(web): validate custom provider base URLs 2026-09-09 12:04:40 +08:00
gengruilin 189ca97ec9 Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 12:04:24 +08:00
Tianyi Cui 08f646da36 Merge pull request #3812 from deepseek-harness/worktree/ci-reliability-master-followup-20260908
test(ci): stabilize asynchronous fixtures and benchmark budgets
2026-09-09 11:52:54 +08:00
gengruilin 454825ca38 Merge origin/master into fix/mcp-pagination-cycles 2026-09-09 11:49:04 +08:00
creatixchu 4b88085c2a style(feedback): refine dialog proportions 2026-09-09 11:41:24 +08:00
Tianyi Cui 2c0596c1ba docs(bench): clarify reviewed request-history limit 2026-09-09 11:29:05 +08:00
pku-xht e8a5e9dfd9 Merge master to restore required CI scheduling 2026-09-09 11:28:38 +08:00
Tianyi Cui ba4b8fddeb test(bench): calibrate hosted cold Agent resume limit 2026-09-09 11:26:48 +08:00
Tianyi Cui 2567d35cdf docs(bench): synchronize request-history limit 2026-09-09 11:26:48 +08:00
Tianyi Cui 843029448a test(code-runtime): allow measured worker initialization time 2026-09-09 11:26:48 +08:00
Tianyi Cui d1b1488ed3 test(session): await durable live writes with owned teardown 2026-09-09 11:26:48 +08:00
Tianyi Cui bd24f8a17f test(app-boot): control user-patch HMR event delivery 2026-09-09 11:26:48 +08:00
Tianyi Cui dc14f5e4f9 test(web): await resized details geometry through its transition 2026-09-09 11:26:48 +08:00
Tianyi Cui 48007d76f2 test(bench): allow measured request-history runner variation 2026-09-09 11:26:48 +08:00
Tianyi Cui fd5d68586f test(web): wait for model selection menu closure 2026-09-09 11:26:48 +08:00
Yichen Jiang 2044d2906b Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages
# Conflicts:
#	apps/web/tsconfig.json
#	tsconfig.host.json
2026-09-09 11:09:16 +08:00
Turtle 9606b69545 Merge origin/master into turtle/issue-3760-zh-display-model-copy 2026-09-09 11:02:57 +08:00
Turtle 43846e7032 Merge pull request #3783 from deepseek-harness/turtle/weighted-approval
feat: enforce weighted pull request approvals
2026-09-09 11:00:39 +08:00
Turtle 9da63ca94d Merge pull request #3789 from deepseek-harness/turtle/issue-3757-settings-accessible-name
fix(client): name collapsed Settings trigger
2026-09-09 10:56:30 +08:00
creatixchu 6207453d36 Merge remote-tracking branch 'origin/master' into worktree/issue-3515-feedback-dialog 2026-09-09 10:44:25 +08:00
pku-xht ad93e2f602 fix(auto-review): preserve nested trajectory errors and optional peer 2026-09-09 10:43:56 +08:00
creatixchu d692289533 style(feedback): compact the dialog to a 540px card with lighter chips and a grown detail box 2026-09-09 10:42:06 +08:00
imccyu c57fb6f7ef Merge pull request #3798 from deepseek-harness/worktree-sidebarfeat
feat(client): document previews — HTML, PDF, Markdown and code
2026-09-09 10:41:39 +08:00
creatixchu 1094431de9 docs(feedback): document the FeedbackSurface fields for the export-jsdoc gate 2026-09-09 10:33:09 +08:00
creatixchu a25e5fd8a9 test(sdk): refresh the feedback expectations on the V3 base and restore the frozen v2 generation 2026-09-09 10:27:15 +08:00
creatixchu 3e53a08f36 fix(feedback): keep the category tuple out of types.ts, retire the toast with its entry, and answer the review 2026-09-09 10:27:12 +08:00
creatixchu e6ff3290cf Merge remote-tracking branch 'origin/master' into worktree/composer-plus-menu
Takes the merged slash-command localization (#3557): its zh description
copy replaces this branch's, its contribution description() callback and
HOST_DESCRIPTION_KEYS are superseded by the label/description/icon
contribution face and presentation.ts, and its zh menu golden and
locale-switch test are refreshed for the sectioned menu.
2026-09-09 10:25:10 +08:00
creatixchu f2f9700de2 feat(feedback): open one feedback dialog from /feedback and Dislike, with categories and a toast
Dislike and a bare /feedback open the Session's feedback dialog, which
collects one of seven fixed categories and an optional description and
records through the messageFeedback put or the new sessionFeedback
Remote; Like records at once. A recorded rating shows the filled glyph
and every submission is acknowledged by a toast. feedback/record and
MessageFeedbackItem carry the optional category; the note popover, its
layout scenario, and the clearNote/clear verbs are removed. ui-commands
gains the action command kind the /feedback decoration runs as.

Closes #3515
2026-09-09 10:20:24 +08:00
imccyu 64560eaff1 fix(client): remove stale preview contracts 2026-09-09 10:12:46 +08:00
imccyu 93b30b9765 fix(client): narrow document preview contracts 2026-09-09 10:12:46 +08:00
imccyu be58ae0577 test(web): await sidebar collapse before snapshots 2026-09-09 10:12:46 +08:00
imccyu 4d07fbcb85 fix(client): address document preview review feedback 2026-09-09 10:12:46 +08:00
imccyu ce28b0672f fix(client): simplify PDF observer cleanup 2026-09-09 10:12:46 +08:00
imccyu 39872bafe6 docs(client): record sidebar guide translation 2026-09-09 10:12:45 +08:00
imccyu c78fdb583a fix(ci): refresh document preview metadata 2026-09-09 10:12:45 +08:00
imccyu 693621dcad fix(client): streamline continuous PDF previews 2026-09-09 10:12:41 +08:00
imccyu dcfd8c299d refactor(client): rename document preview package 2026-09-09 10:12:40 +08:00
imccyu 5c283d27e2 fix(client): allow outside previews and streamline document readers 2026-09-09 10:12:39 +08:00
imccyu 4321766ead feat(client): add extensible document previews and sidebar defaults 2026-09-09 10:12:38 +08:00
pku-xht c0348f7bfc Merge master to restore required CI scheduling 2026-09-09 08:35:58 +08:00
pku-xht 41fbde8034 fix(permission): revoke stale Auto choices and simplify review paths 2026-09-09 08:08:10 +08:00
Yichen Jiang e9b3f80f0b Merge pull request #3724 from deepseek-harness/worktree/pi-ai-settings-recovery
fix(llm-pi-ai): keep catalog-invalid settings repairable
2026-09-09 06:37:54 +08:00
Yichen Jiang d80891921b Merge remote-tracking branch 'origin/master' into worktree/pi-ai-settings-recovery
# Conflicts:
#	packages/llm/llm/README.i18n.yaml
#	packages/llm/llm/README.md
#	packages/llm/llm/README.zh.md
2026-09-09 06:18:13 +08:00
pku-xht 0eb0e8c770 Merge master to restore required CI for Auto review 2026-09-09 05:25:20 +08:00
pku-xht 9cfb120a63 merge: reconcile experimental Auto review implementation 2026-09-09 04:38:55 +08:00
pku-xht 55e53907ab feat(permission): add experimental Auto review 2026-09-09 04:36:19 +08:00
imccyu ab904f4b6b Merge pull request #3813 from deepseek-harness/worktree-nodeaddworkflow
ci(native): consolidate platform compatibility jobs
2026-09-09 02:45:46 +08:00
imccyu 063165a800 ci(native): consolidate platform compatibility jobs 2026-09-09 02:31:43 +08:00
fz 0ba6c25807 Merge pull request #3510 from deepseek-harness/feat/remove-str-replace-editor-from-minimal
feat(minimal): make editor opt-in and align bash output
2026-09-09 02:10:26 +08:00
fz 452fe6de55 Merge remote-tracking branch 'origin/master' into feat/remove-str-replace-editor-from-minimal 2026-09-09 01:49:09 +08:00
fz 887c1efd7a ci(npm): expose failed install HTTP diagnostics 2026-09-09 01:48:54 +08:00
fz 0f6e7442d7 perf(session): bundle migration verifier dependencies 2026-09-09 01:48:27 +08:00
Yudong Han 6635cddaa4 Merge pull request #3569 from deepseek-harness/fs/stream-bytes
feat(web): declare workspace deliveries and open source files
2026-09-09 01:33:20 +08:00
fz 40c9766a8d Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-09 01:22:09 +08:00
Yudong 228f3aef83 test(ci): stabilize completion waits and Session fixtures 2026-09-09 01:15:31 +08:00
yudshj 72549e91a0 test(present): pass the event to delivery matcher 2026-09-09 01:15:31 +08:00
yudshj cd2b9ca04a refactor(present): restore existing conversation coordinate handling 2026-09-09 01:15:31 +08:00
yudshj 207dce0bee docs(present): refresh module graph after removing attachments 2026-09-09 01:15:31 +08:00
yudshj c9038b4b31 refactor(present): declare and open workspace source files 2026-09-09 01:15:31 +08:00
yudshj 387e437af4 fix(web): use the preview-compatible stream promises entry 2026-09-09 01:15:31 +08:00
yudshj 6d8a400d79 fix(web): open delivered snapshots in the default application 2026-09-09 01:15:31 +08:00
yudshj 37d27fcdf9 feat(web): present immutable file deliveries with download cards 2026-09-09 01:15:31 +08:00
imccyu 4d831f2d14 Merge pull request #3771 from deepseek-harness/worktree-fixreactpeer
fix(web): move client dependencies to devDependencies
2026-09-09 01:15:27 +08:00
imccyu bfe82d4c2d Merge pull request #3784 from deepseek-harness/worktree-leftsideslot
feat(web): add panellist to left-sidebar
2026-09-09 00:59:11 +08:00
Dudu-0223 a4176ef851 Merge pull request #3674 from deepseek-harness/feat/parent-subagent-catalog-foundation
feat(subagent): record and observe parent-owned child catalogs
2026-09-09 00:37:47 +08:00
imccyu f39a37a523 fix(client): keep browser dependencies out of production installs 2026-09-09 00:30:37 +08:00
imccyu f3d7ef2b95 docs(client): define browser third-party build inputs 2026-09-09 00:30:36 +08:00
imccyu 58144ccf78 test(web): await command and settings readiness before assertions 2026-09-09 00:26:41 +08:00
imccyu e807447288 fix(client): re-read navigation cancellation after preparation 2026-09-09 00:26:41 +08:00
imccyu 713b1aacad docs(client): synchronize panel navigation APIs and contracts 2026-09-09 00:26:40 +08:00
imccyu 385f651c5f fix(test): release default slot runtime sources symmetrically 2026-09-09 00:26:36 +08:00
imccyu 8715fd2a87 fix(client): preserve the latest panel and session navigation 2026-09-09 00:26:36 +08:00
imccyu bc840b483e feat(client): simplify and exercise sidebar panel entries 2026-09-09 00:26:35 +08:00
imccyu 092d16d7ab style(client): satisfy panel lint rules 2026-09-09 00:26:35 +08:00
imccyu 9736136d95 fix(client): leave global panel entries empty by default 2026-09-09 00:26:34 +08:00
imccyu fbbcbf33fe docs(client): document global main-panel extension points 2026-09-09 00:26:34 +08:00
imccyu f95e781f3d test(client): migrate panel navigation fixtures and snapshots 2026-09-09 00:26:29 +08:00
imccyu fe460f7fd1 test(client): support panel selection in slot runtime 2026-09-09 00:26:29 +08:00
imccyu 581803bf57 feat(client): add global sidebar panels 2026-09-09 00:26:28 +08:00
Yichen Jiang e5feedd14f feat(llm-deepseek-messages): support in-history system prompt updates 2026-09-09 00:05:41 +08:00
fz b42b5b5e94 fix(ci): align minimal prompt-history expectations with shell-only turns 2026-09-08 23:57:15 +08:00
Yichen Jiang 1f88b9cd8f chore(llm-deepseek-messages): align version with master 2026-09-08 23:44:14 +08:00
Yichen Jiang d21ee05ebb Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-08 23:38:48 +08:00
gengruilin 594305ce19 fix(mcp): reject repeated tool discovery cursors
Refs: https://github.com/deepseek-ai/deepseek-harness/discussions/3660
2026-09-08 23:33:23 +08:00
fz 25d675fcf3 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-08 23:31:51 +08:00
Dudu-0223 4977ccc965 chore(chunked-list): follow master release version 2026-09-08 23:29:58 +08:00
fz 3c05dcb537 fix(ci): refresh persistent bash fixtures and synchronize LSP lifecycle test 2026-09-08 23:29:03 +08:00
Dudu-0223 5ab91467e2 test(cli): include catalog facts in profile expectations 2026-09-08 23:28:50 +08:00
Dudu-0223 79dda6cf45 test(subagent): align catalog snapshots with Session V3 2026-09-08 23:28:50 +08:00
Dudu-0223 e2fa69f5e2 test(snapshot): bind sibling roles to catalog publication order 2026-09-08 23:28:50 +08:00
Dudu-0223 9f69a7fb16 refactor(subagent): extract persistent chunked list utility 2026-09-08 23:28:50 +08:00
Dudu-0223 661135fe29 fix(subagent): reject invalid catalogs and preserve snapshot event order 2026-09-08 23:28:50 +08:00
Dudu-0223 ba731d1c9e fix(subagent): handle rejected runs after catalog failure 2026-09-08 23:28:50 +08:00
Dudu-0223 867c47c415 docs: drop unrelated projection README whitespace changes 2026-09-08 23:28:50 +08:00
Dudu-0223 59623c8ff4 docs(subagent): refresh event declaration references 2026-09-08 23:28:50 +08:00
Dudu-0223 0274a6dd75 fix(subagent): preserve catalog augmentation in declarations 2026-09-08 23:28:50 +08:00
Dudu-0223 924282cbcd fix(subagent): retain catalog event types in host exports 2026-09-08 23:28:50 +08:00
Dudu-0223 532e712c7b refactor(subagent): expose catalog through existing projection views 2026-09-08 23:28:50 +08:00
Dudu-0223 3dcabbba44 refactor(session-query): return all projection states by default 2026-09-08 23:28:50 +08:00
Dudu-0223 e3cb6dd352 refactor(session): limit hydration views to all or none 2026-09-08 23:28:50 +08:00
Dudu-0223 80e34ce709 perf(subagent): list parent catalog in event order 2026-09-08 23:28:50 +08:00
Dudu-0223 99462aef98 fix(subagent): preserve catalog citations and skip unrequested views 2026-09-08 23:28:50 +08:00
Dudu-0223 399d7313c4 test(session): cover historical catalog comparison failures 2026-09-08 23:28:50 +08:00
Dudu-0223 2db4bdd31d feat(subagent): record and observe parent-owned child catalogs 2026-09-08 23:28:50 +08:00
fz 94ff82f57b Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-08 23:22:25 +08:00
Turtle f830cfd66a Merge remote-tracking branch 'origin/master' into turtle/weighted-approval
# Conflicts:
#	.github/review-ownership/README.md
2026-09-08 21:06:57 +08:00
Turtle c2a5540ddb fix(web): distinguish plugin configuration instances 2026-09-08 21:05:37 +08:00
Turtle 555c04bdfc fix(web): localize Chinese model copy 2026-09-08 21:05:31 +08:00
Turtle 3c21d4c28c fix(client): name collapsed Settings trigger 2026-09-08 20:38:10 +08:00
Turtle c4855581c3 fix: preserve weighted approval review semantics 2026-09-08 20:19:29 +08:00
fz 1f1537914f feat(minimal): align persistent bash output with the one-shot shell contract
The minimal compositions serve the same models that trained against a
one-shot `bash` tool; three surface differences remained visible to them.

- Every settled command now appends `[Command finished with exit code N]`,
  including exit code 0, instead of `[exit code: N]` only on failure. A
  timeout appends `[Command timed out or OOM]` after the partial output.
- Trailing newlines are all dropped before the status trailer rather than
  only the last one.
- The `bash` description in the `minimal` preset and the `sdk-minimal`
  bundle replaces the two fixed network lines with a single environment-
  dependent line; the pwsh description is unchanged.

Tests, READMEs, the web minimal-preset schema snapshot, and the packaged
Python minimal model-visible snapshot follow the new wording. The
standalone `persistent-tools` SDK replay fixture keeps its own copy of the
old description since it is a per-launch customization, not the profile.
2026-09-08 19:48:21 +08:00
Turtle 4550bb6e69 feat: enforce weighted pull request approvals 2026-09-08 19:08:08 +08:00
07akioni 8874a23b9e feat: 加速 mac sign 2026-09-08 17:44:05 +08:00
07akioni 8e4d3bab9f feat: 优化打包方式、启动速度 2026-09-08 16:39:17 +08:00
Yichen Jiang da1bc2523a Merge branch 'master' into worktree/pi-ai-settings-recovery 2026-09-08 16:28:15 +08:00
Yichen Jiang dd0fe585dd Merge remote-tracking branch 'origin/master' into worktree/pi-ai-settings-recovery 2026-09-08 16:07:59 +08:00
Yichen Jiang aa3a1c2b4a Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages
# Conflicts:
#	packages/llm/llm-deepseek/README.i18n.yaml
#	packages/llm/llm-deepseek/README.md
#	packages/llm/llm-deepseek/README.zh.md
2026-09-08 15:58:22 +08:00
Yichen Jiang 914dec6fea test(web): refresh Messages composer and timezone snapshots 2026-09-08 15:55:07 +08:00
Yichen Jiang e52ffcea87 Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-08 15:49:35 +08:00
Yichen Jiang c9a702a267 Merge remote-tracking branch 'origin/master' into worktree/pi-ai-settings-recovery 2026-09-08 15:42:48 +08:00
Yichen Jiang dad135dd41 Merge remote-tracking branch 'origin/master' into worktree/pi-ai-settings-recovery 2026-09-08 15:42:25 +08:00
creatixchu 8ccf82cbf2 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-08 15:22:06 +08:00
creatixchu 4779f054b6 refactor(compaction): own image offload as a compaction executor
Move the IMAGE_OFFLOAD_REQUIRED recovery out of llm-retry into the new
@deepseek-ai/dsh-compaction-image-offload package, a sibling of the
tool-result pruner: image offload is compaction in another capacity
dimension. Each replaced node is priced through the seam's compaction/prune
shadow-price protocol before its marked copy lands. llm-retry returns to
master; the base bundle mounts the executor.
2026-09-08 15:10:16 +08:00
creatixchu 6b673aedf5 docs(llm): drop the last watermark wording from llm-streaming and the retry test title 2026-09-08 14:08:47 +08:00
fz 61cd1a3f91 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-08 13:51:47 +08:00
creatixchu c755e3e501 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-08 13:50:32 +08:00
Yichen Jiang da313e4fdc fix(llm-pi-ai): preserve actionable model diagnostics 2026-09-08 13:35:09 +08:00
Huanqi Cao a39d309187 fix(directory-picker-native): foreground-activate the Win32 folder dialog via a synthesized Alt press 2026-09-08 13:10:03 +08:00
creatixchu c75e100893 fix(client): share the shield contour and describe the menu glyphs by shape
The permission selector and IconShieldOutline16 now read one exported
SHIELD_OUTLINE_PATH; the icon-set header no longer claims every glyph is
filled; the three menu glyph docs name what they draw and where they are
used; the compact ring matches the set's extent; the icon pin asserts the
three names.
2026-09-08 12:47:21 +08:00
creatixchu 5b1bb021cf feat(web): group, localize, and re-present the composer command menu
The + button and a typed / open one menu with an Add section (File, Goal,
Plan, Feedback) and a Commands section (Compact, Permission, Model, Export)
in usage order; rows carry a glyph, a localized title and description, and
the command name as an alias where the localized title differs. Built-in
Host rows get their face from the ui-commands dictionaries when the catalog
description is the canonical English text. A menu pick fills the locale's
claim token (/计划 ) while the submission executes the catalog name, and a
typed localized token resolves on Space and Enter in every locale. The File
row is a new action contribution kind that opens the composer's file picker
through the scoped slash/input-pick-files event; the paperclip button is
removed and the + button reads "Add files or run commands".
2026-09-08 12:28:40 +08:00
creatixchu 2c9eda5b6f feat(client): add plan, compact, and shield glyphs for the composer menu 2026-09-08 12:05:06 +08:00
Yichen Jiang 8dc30d270c test(llm-pi-ai): cover recovery diagnostics and document repair limits 2026-09-08 12:04:18 +08:00
creatixchu 6329c53a37 refactor(llm-retry, session): offload images by surface replacement instead of a watermark event
Drop the image/offload event and every session change with it. An adapter's
IMAGE_OFFLOAD_REQUIRED failure now makes dsh-llm-retry replace the surface
nodes carrying the named oldest images with copies whose blocks are marked
offloaded, through the same surfaceOp: replace compaction uses, then retry the
step. ImageBlock.offloaded becomes a durable field the routes project and the
token meter prices; the session core, its event map, derivation, and
validation return to master.
2026-09-08 12:00:35 +08:00
Yichen Jiang fa48b101d2 test(web): align Messages snapshot with Sidebar 2026-09-08 11:33:58 +08:00
fz ad05ed6ad8 docs(minimal): describe current editor availability 2026-09-08 11:33:24 +08:00
Yichen Jiang db63b2be7f Merge remote-tracking branch 'origin/master' into worktree/llm-deepseek-messages 2026-09-08 11:30:28 +08:00
Yichen Jiang 5bc792f629 refactor(llm): keep diagnostics in configurable provider metadata 2026-09-08 10:55:53 +08:00
Yichen Jiang a12d30df2a refactor(llm-pi-ai): reuse settings validation for startup recovery 2026-09-08 10:47:05 +08:00
creatixchu 52386ec6fb test(llm-retry): follow agent loop testkit dependencies 2026-09-08 10:46:02 +08:00
creatixchu 73cee076fb Merge remote-tracking branch 'origin/master' into codex/pr-3440-resync-20260908 2026-09-08 10:37:36 +08:00
Yichen Jiang 642b538fe0 Merge remote-tracking branch 'origin/master' into worktree/pi-ai-settings-recovery 2026-09-08 10:34:14 +08:00
Yichen Jiang fe03b0686d fix(llm-pi-ai): keep catalog-invalid settings repairable 2026-09-08 10:28:45 +08:00
fz 707d0e7e52 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-07 23:25:03 +08:00
fz b92cc94122 test(minimal): cover documented editor opt-in 2026-09-07 22:56:48 +08:00
Yichen Jiang 74ef3f0510 fix(llm): degrade unusable Messages replay metadata 2026-09-07 21:32:06 +08:00
Yichen Jiang c258d3cf4e feat(web): use Messages for the default DeepSeek provider 2026-09-07 20:57:23 +08:00
creatixchu e44eb89065 test(llm-retry): cover the surface node that derives no message 2026-09-07 20:19:30 +08:00
creatixchu 523ae08a2d refactor(session, llm): cut the image offload surface to its callers
Drop Session.imageOffloadWatermark(), which no production code read once
recovery moved to llm-retry; shorten the append and seed validators; keep
offloadedImagePrefixCount and the represented-bytes arithmetic private to
requiredImageOffload, their only caller.
2026-09-07 20:18:56 +08:00
Yichen Jiang 57c6e7e9b2 fix(web): remove Messages endpoint helper text 2026-09-07 20:09:16 +08:00
Yichen Jiang 23fa53ecbc Merge master release and align Messages package version 2026-09-07 19:55:51 +08:00
fz 110db78707 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-07 19:53:33 +08:00
fz 5c98a81a50 test(benchmarks): explicitly compose the SDK file editor 2026-09-07 19:43:49 +08:00
Yichen Jiang d28286938d feat(web): configure DeepSeek Messages beside Chat Completions 2026-09-07 19:39:37 +08:00
fz d3281d0218 docs(minimal): explain editor opt-in configuration 2026-09-07 19:33:48 +08:00
fz 71a291cbc4 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-07 19:28:12 +08:00
Yichen Jiang 34154b6861 feat(llm): add DeepSeek Anthropic Messages adapter 2026-09-07 19:20:29 +08:00
creatixchu ab93bfd90c refactor(session): fold the image offload helpers into surface.ts and index.ts
The watermark marking and position comparison sit next to deriveEventMessage
in surface.ts; the append/seed validation sits with the other event validators
in index.ts and the fold is inline. Only compareImagePositions stays exported.
2026-09-07 17:35:28 +08:00
creatixchu fcb976d3dc refactor(llm): recover IMAGE_OFFLOAD_REQUIRED in llm-retry instead of a new plugin package
Remove the llm-image-offload package and every registration it needed. The
recovery branch lives in dsh-llm-retry, which already sits on the
agent/request-error waterfall: an IMAGE_OFFLOAD_REQUIRED failure advances the
image/offload watermark by the count the adapter names and retries the step
outside the provider retry budget. Drop pre-dispatch planning with it: routes
no longer declare a request-image budget on their model info, and the byte
checks that stay in the DeepSeek and pi-ai adapters are provisional until the
provider reports uncacheable images itself.
2026-09-07 17:03:23 +08:00
creatixchu cea09de58c Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-07 15:58:23 +08:00
creatixchu 8c7b023d6c refactor(llm): move image offload planning into a plugin and trim the change surface
Keep the agent loop unchanged: the new @deepseek-ai/dsh-llm-image-offload
plugin plans the image/offload watermark on agent/pre-step from the routed
model's budget and advances it on agent/request-error when an adapter fails
with IMAGE_OFFLOAD_REQUIRED, returning the retry action the loop already
honours. Drop PreparedLlmCall.imageRequest and its validation, the planner
wrapper, the replay adapter's imageRequestMaxBytes option, and the duplicate
ACP snapshot scenario; share the exact-byte overflow count between the
DeepSeek and pi-ai adapters through requiredImageOffload(). The SDK image
snapshot replays an authored IMAGE_OFFLOAD_REQUIRED attempt instead of a
replay-only budget.
2026-09-07 15:58:20 +08:00
creatixchu 6b3681ae44 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design
# Conflicts:
#	docs/architecture.i18n.yaml
#	docs/architecture.md
#	docs/architecture.zh.md
2026-09-07 14:59:46 +08:00
creatixchu b6cb4609b6 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-07 14:40:05 +08:00
creatixchu 74f380806f test(snapshot): follow persona prefix config 2026-09-07 14:13:01 +08:00
creatixchu e147077526 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-07 14:08:26 +08:00
creatixchu fdbeeddc36 fix(session): tolerate contentless synthetic messages 2026-09-07 13:01:19 +08:00
creatixchu 5e277c2111 test(session): close image offload coverage gaps 2026-09-07 12:51:09 +08:00
creatixchu 57a3701a22 Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-07 12:35:43 +08:00
creatixchu 60f9d9b936 Merge master and address image offload review 2026-09-07 12:27:48 +08:00
pku-xht fdeeacef74 fix(web): revoke withdrawn permission choices 2026-09-06 09:32:09 +08:00
pku-xht 197161e660 fix(permission): simplify Auto identity and teardown 2026-09-06 07:14:09 +08:00
pku-xht 0861ee11f7 test(permission): cover Auto lifecycle and inheritance boundaries 2026-09-06 05:29:42 +08:00
pku-xht cdce3a9137 Merge master to restore required CI scheduling 2026-09-06 04:34:20 +08:00
pku-xht 2505fcaf97 fix(auto-review): align risk policy and core certification 2026-09-06 04:16:13 +08:00
fz 40539e9002 Merge origin/master into feat/remove-str-replace-editor-from-minimal 2026-09-06 00:58:49 +08:00
pku-xht 492c37494a test(web): align Auto denial goldens with composer 2026-09-05 06:22:11 +08:00
pku-xht c7fbe079f0 fix: align delegated preset snapshots 2026-09-05 05:57:39 +08:00
pku-xht 344432de76 Merge commit '746fc39c75c77eb7663f916002ae9e3fd9827a71' into auto-review-per-call
# Conflicts:
#	docs/config-catalog.i18n.yaml
#	docs/config-catalog.md
#	docs/config-catalog.zh.md
#	docs/event-producer-consumer.i18n.yaml
#	docs/event-producer-consumer.md
#	docs/event-producer-consumer.zh.md
#	packages/api/remotes/src/client/index.ts
#	packages/client/ui-conversation/src/client/locales.ts
#	packages/client/ui-conversation/src/client/skeleton/InputBar.tsx
#	packages/client/ui-conversation/tests/input-bar.client.spec.tsx
#	packages/extensions/cordis-client-runner/src/client/slot-catalog.ts
2026-09-05 03:46:30 +08:00
pku-xht 3993d36c69 fix(auto-review): stabilize allow response protocol 2026-09-04 15:45:54 +08:00
pku-xht 5b6462f51c Merge commit 'c8e5c4c3a2315079b3ba0b26bc6e6c6b1edd204b' into auto-review-per-call 2026-09-04 15:03:41 +08:00
pku-xht 3879243d62 docs(auto-review): clarify teardown ordering 2026-09-04 14:57:21 +08:00
pku-xht 91b504e63e ci(e2e): isolate auto certification 2026-09-04 13:20:05 +08:00
pku-xht e6863256f7 ci(e2e): serialize shared-key real-model tests 2026-09-04 13:01:31 +08:00
pku-xht 8481e02b36 fix(auto-review): reinforce reviewer output contract 2026-09-04 12:46:25 +08:00
pku-xht 11c4fa1b0a fix(permissions): harden auto disposal migration 2026-09-04 12:21:56 +08:00
pku-xht dae5844f93 fix(auto-review): scope logged calls by step 2026-09-04 12:10:10 +08:00
pku-xht ca8d65e5c4 docs(auto-review): avoid deterministic output claim 2026-09-04 09:56:45 +08:00
pku-xht 5a99cd9a47 fix(auto-review): stabilize reviewer decisions 2026-09-04 09:27:16 +08:00
pku-xht 96c29c2e6a docs: refresh auto review event listeners 2026-09-04 08:47:30 +08:00
pku-xht 0df80cdf41 fix(auto-review): carry lifecycle cancellation to dispatch 2026-09-04 08:08:52 +08:00
pku-xht a3d00cebc0 Merge commit '0d361e6eede9e8a739e776949aa7d5e3de4472da' into auto-review-per-call
# Conflicts:
#	.agents/notes/implemented/feature/2026-07-25-subagent-policy-inheritance.i18n.yaml
#	.agents/notes/implemented/feature/2026-07-25-subagent-policy-inheritance.md
#	.agents/notes/implemented/feature/2026-07-25-subagent-policy-inheritance.zh.md
#	docs/event-producer-consumer.i18n.yaml
#	docs/event-producer-consumer.md
#	docs/event-producer-consumer.zh.md
#	docs/persistence-catalog.i18n.yaml
#	docs/persistence-catalog.md
#	docs/persistence-catalog.zh.md
#	packages/extensions/tool-cordis/src/api-catalog.ts
#	snapshots/session/cordis-inspect-jsdoc/session.v2.jsonl
#	snapshots/web/ptc-round/session.v2.jsonl
2026-09-04 07:10:24 +08:00
pku-xht d381a3c6f4 feat(auto-review): enforce risk-classified review 2026-09-04 06:46:20 +08:00
fz 8d06ebb5ac Revert "refactor: remove str_replace_editor"
This reverts commit d12b1146baa326e6facd3b10f81ae11db35ffd86.
2026-09-03 21:05:38 +08:00
fz 79b3d7a594 refactor: remove str_replace_editor 2026-09-03 17:02:37 +08:00
creatixchu 345b5cdc6f feat(session, llm): durable image offload watermark
Record the image offload point as the core image/offload session event so the
derived surface, every adapter, and the token meter share one offloaded set.
The agent loop plans the advance from the route's declared request-image
budget before dispatch; an adapter whose exact accounting still overflows
fails with IMAGE_OFFLOAD_REQUIRED and the loop advances by the named count.
2026-09-03 16:29:51 +08:00
fz e9f8641e76 Merge remote-tracking branch 'origin/master' into feat/remove-str-replace-editor-from-minimal
# Conflicts:
#	.agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.i18n.yaml
#	.agents/notes/implemented/architecture/2026-08-23-python-sdk-dsh-profile-runtime.i18n.yaml
#	.agents/notes/implemented/architecture/2026-08-23-python-sdk-windows-x64-runtime.i18n.yaml
2026-09-03 16:05:56 +08:00
fz 63795eaa5c feat(minimal): remove str_replace_editor from minimal profiles 2026-09-03 16:03:59 +08:00
creatixchu abb2b31dbe Merge remote-tracking branch 'origin/master' into worktree/3041-image-offload-log-design 2026-09-03 14:40:29 +08:00
creatixchu c1b1cd2984 docs: propose a durable image offload watermark 2026-09-03 14:40:24 +08:00
pku-xht 681b0b66a9 fix(auto-review): align review and teardown boundaries 2026-09-03 08:07:42 +08:00
pku-xht 92cb99e9e9 test(snapshot): canonicalize Auto review PTC fixture 2026-09-03 06:39:10 +08:00
pku-xht 1ec5f21e82 fix(permission): tighten Auto review validation 2026-09-03 05:33:34 +08:00
pku-xht 64b05dacdd Merge commit '7dd92d1c8c5a6b1cf47ac677026fb08296abb809' into auto-review-per-call
# Conflicts:
#	apps/web/tests/shipped-composition.e2e.ts
#	docs/module-graph.i18n.yaml
#	docs/module-graph.md
#	docs/module-graph.zh.md
#	docs/persistence-catalog.i18n.yaml
#	docs/persistence-catalog.md
#	docs/persistence-catalog.zh.md
#	packages/client/ui-conversation/src/client/apply.ts
#	packages/client/ui-conversation/src/client/contract/slots.ts
#	packages/client/ui-conversation/src/client/skeleton/InputBar.tsx
#	packages/client/ui-conversation/tests/input-bar.client.spec.tsx
#	packages/client/ui-conversation/tsconfig.json
#	packages/client/ui-tool/README.i18n.yaml
#	packages/client/ui-tool/README.md
#	packages/client/ui-tool/README.zh.md
#	packages/client/ui-tool/src/client/tool/ToolCallTree.tsx
#	packages/client/ui-tool/src/client/tool/components/ToolRow.tsx
#	packages/extensions/tool-cordis/src/api-catalog.ts
2026-09-03 02:44:29 +08:00
pku-xht 535d7eab3a feat(auto-review): enforce per-call authorization 2026-09-03 02:17:30 +08:00
creatixchu 1516c6018d docs: propose logging each request's image projection 2026-09-02 14:53:45 +08:00
pku-xht 3f610869b6 Merge commit '05380955d3d59a536c6c4f12def7c1586daaa613' into auto-review-per-call
# Conflicts:
#	docs/subsystems/session-projection.i18n.yaml
#	docs/subsystems/session-projection.md
#	docs/subsystems/session-projection.zh.md
#	packages/client/ui-conversation/tests/input-bar.client.spec.tsx
#	packages/client/ui-permission-presets/README.i18n.yaml
#	packages/client/ui-permission-presets/README.md
#	packages/client/ui-permission-presets/README.zh.md
#	packages/client/ui-permission-presets/tests/browser-plugin.client.spec.ts
#	packages/client/ui-tool/README.i18n.yaml
#	packages/client/ui-tool/README.md
#	packages/client/ui-tool/README.zh.md
#	packages/client/ui-tool/src/client/tool/components/ToolRow.tsx
#	packages/client/ui-tool/src/client/tool/toolviews/bash-sample.tsx
#	packages/client/ui-tool/tests/tool-row.client.spec.tsx
#	packages/extensions/tool-cordis/src/api-catalog.ts
#	packages/session/session-projection/README.i18n.yaml
#	packages/session/session-projection/README.md
#	packages/session/session-projection/README.zh.md
#	packages/session/session-projection/src/index.ts
#	packages/session/session-projection/tests/registry.spec.ts
2026-08-31 22:07:28 +08:00
pku-xht 02f20b7759 Merge commit '40ef442e90e9d9056982edc605db476a2b760ecd' into auto-review-per-call
# Conflicts:
#	apps/web/tests/shipped-composition.e2e.ts
#	docs/config-catalog.i18n.yaml
#	docs/config-catalog.md
#	docs/config-catalog.zh.md
#	docs/event-producer-consumer.i18n.yaml
#	docs/event-producer-consumer.md
#	docs/event-producer-consumer.zh.md
#	docs/module-graph.i18n.yaml
#	docs/module-graph.md
#	docs/module-graph.zh.md
#	docs/persistence-catalog.i18n.yaml
#	docs/persistence-catalog.md
#	docs/persistence-catalog.zh.md
#	packages/client/ui-conversation/src/client/locales.ts
#	packages/client/ui-conversation/src/client/skeleton/PermissionSelect.tsx
#	packages/client/ui-conversation/tests/input-bar.client.spec.tsx
#	packages/client/ui-permission-presets/README.i18n.yaml
#	packages/client/ui-permission-presets/README.md
#	packages/client/ui-permission-presets/README.zh.md
#	packages/client/ui-permission-presets/src/client/index.ts
#	packages/client/ui-permission-presets/src/client/locales.ts
#	packages/client/ui-permission-presets/src/client/presentation.ts
#	packages/client/ui-permission-presets/tests/browser-plugin.client.spec.ts
#	packages/core/tools/src/ptc.ts
#	packages/interaction/permission-presets/src/index.ts
2026-08-31 17:25:58 +08:00
pku-xht 07e4ceb011 fix(web): localize Auto review descriptions 2026-08-31 14:18:51 +08:00
pku-xht 068c6ca3ff test(ui-tool): trim redundant Auto denial matrix 2026-08-29 06:57:15 +08:00
pku-xht e6ae1bc751 refactor(permission): simplify Auto review ownership 2026-08-29 06:24:58 +08:00
pku-xht 64fa60d3d4 fix(permission): close auto review gaps 2026-08-29 05:01:23 +08:00
pku-xht 1912573bde Merge commit '1ffc132ea51cfa57a86ae124ca3de6e650337c8c' into auto-review-per-call
# Conflicts:
#	docs/config-catalog.i18n.yaml
#	docs/config-catalog.md
#	docs/config-catalog.zh.md
#	docs/persistence-catalog.i18n.yaml
#	docs/persistence-catalog.md
#	docs/persistence-catalog.zh.md
#	docs/subsystems/permission-presets.i18n.yaml
#	docs/subsystems/permission-presets.md
#	docs/subsystems/permission-presets.zh.md
#	knip.json
#	packages/extensions/tool-cordis/src/api-catalog.ts
#	packages/interaction/permission-presets/README.i18n.yaml
#	packages/interaction/permission-presets/README.md
#	packages/interaction/permission-presets/README.zh.md
#	packages/interaction/permission-presets/src/index.ts
#	packages/interaction/permission-presets/tests/permission-presets.spec.ts
2026-08-29 02:08:28 +08:00
pku-xht b3b7e00449 feat(permission): add Auto review mode 2026-08-29 00:56:21 +08:00
4570 changed files with 813857 additions and 56485 deletions

No files matched your search

@@ -0,0 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/archived/architecture/2026-09-02-durable-image-offload.md
2026-09-02-durable-image-offload.md: db03e9f52499c1790b3a0c03853f3aea23315bec
2026-09-02-durable-image-offload.zh.md: 2b6d0a4aa65f9a39b88888506f7e85b4cfff48cf
@@ -0,0 +1,56 @@
# Agent Note: Durable image offload by surface replacement
Status: implemented
Archived: 2026-09-10
English | [中文](2026-09-02-durable-image-offload.zh.md)
## Problem
Request-size image offload was recomputed from scratch on every request. Each route collected every image occurrence on the derived surface, oldest first, and once the accumulated bytes exceeded its budget it rounded the excess up to a whole removal quantum and replaced that many oldest occurrences with placeholder text, per the [unified image request pipeline](../feature/2026-08-20-unified-image-request-pipeline.md). Nothing remembered where the previous request stopped; the prefix was stable only because the arithmetic over an append-only history repeated itself.
That stability failed wherever the arithmetic inputs moved. The [Files inline fallback](../../archived/bug-fix/2026-08-21-deepseek-files-inline-fallback.md) rebuilt a request under a 20 MiB inline budget with a 10 MiB quantum, offloading far more images for that one request, and the next request in file mode brought them back. The pi-ai route used a quantum of one byte, so its prefix moved on almost every request. Compaction lowered the total and returned previously offloaded images. A route switch moved every step boundary. Each move changed the model-visible prefix and invalidated the provider cache prefix.
The same recomputation broke the repository invariant that model-visible input is reconstructable from the session log. Which representation was dispatched, the exact derived request-version byte lengths, and the route budgets and quanta were runtime or configuration facts that never entered the log; `request/header` records call config, system prompt, and tools only. Provider usage anchors token totals and cannot recover the image set, and the [route-priced estimate](../../archived/feature/2026-08-24-route-priced-image-request-pressure.md) documented that it did not reproduce the fallback budget. No consumer could pair a logged assistant response with the image set its request carried.
## Decision
The offload of an image occurrence is a durable surface fact recorded the way compaction records its reductions: a `surfaceOp: replace` node.
**Marked copies on the surface.** `ImageBlock` gains `offloaded?: true`. An offloaded occurrence lives in a replacement event of the same type as the node that carried it (`user/message` or `tool/result`) whose content is a copy of the original with those blocks marked; `sourceEventSeqs` points at the replaced node. The session core, its event map, its derivation, and its validation are unchanged. `deriveMessages()` sends the marked block; serialization renders every marked block as `offloadedImageText` with the currently resolved access path and prepares only retained occurrences.
**Only advances.** A replacement never reverts. When a budget grows, a route changes, or compaction lowers the total, the marked copies stay, so the model-visible prefix and the provider cache prefix move only forward.
**Adapters project, never decide.** An image-capable route enforces an `LlmImageRequestBudget` (`representation`, `maxBytes`, `maxImages`, both quanta) over the retained occurrences' exact request-version bytes. When they still exceed the budget, in file mode, under the inline fallback's tighter budget, or under the pi-ai bound, the adapter fails the attempt with `IMAGE_OFFLOAD_REQUIRED` and `LlmFailure.offloadImages` naming how many more oldest occurrences must be offloaded, computed with the shared `requiredImageOffload()`. Nothing plans an offload before dispatch.
**Recovery owned by `dsh-compaction-image-offload`.** Image offload is compaction in another capacity dimension: the provider rejects a request, durable history is reduced, the step retries. The executor is a sibling of `compaction-tool-result-pruner` in the compaction group and listens on the `agent/request-error` waterfall. On `IMAGE_OFFLOAD_REQUIRED` it walks the surface in model request order, marks the first `offloadImages` retained occurrences, and for each node that carried one appends the seam's `compaction/prune` shadow price followed by the marked copy, then returns the `retry` action without spending the provider retry budget or logging `llm/retry`. Assistant nodes carry model output, not input images, and are skipped. When nothing remains to offload it delegates and the failure reaches ordinary recovery. The loop re-runs the step over the replaced surface and logs the fresh `request/header` every surface replacement gets; the agent loop is unchanged.
**Token accounting.** `priceImages` receives the surface's `ImageBlock`s and prices a marked one as its placeholder text; the DeepSeek and replay pricing no longer reproduce any offload arithmetic. The meter needs no new state: the `compaction/prune` event and the replacement reprice the node like a tool-result prune.
**Other consumers.** Compaction reconstructs each selected event through `deriveEventMessage()` and sees the marks. Resume, fork, and replay reproduce the surface from the log. Text-only routes keep their separate whole-history substitution.
## Alternatives considered
**Keep recomputing the offload point per request.** Stable only while the arithmetic inputs held still; the inline fallback, the pi-ai quantum, compaction, and route switches all moved the prefix, and no consumer could reconstruct a historical request's image set.
**Record each request's projection outcome as a log-only event.** Restores reconstructability but not stability: the recorded outcome is not a decision input, so every oscillation still happens and the log merely documents it, with two sources of truth for the offloaded set that can disagree.
**Log the full projected request body.** Everything except the offload decision is already derivable; repeating the history per request grows the log quadratically to record one position.
**A log-only `image/offload` watermark event applied by session derivation.** An earlier revision recorded the offload point as a position (event seq plus block path) and had `Session.deriveMessages()` mark every occurrence at or before it. That gave the session a second mechanism for changing model-visible history next to `surfaceOp: replace`, with its own validation, fold, cache invalidation, and required-on-read event, and placed the logic in the core instead of on the plugin that decides. Issue #3041 asks that a permanent eviction use a surface-changing event rather than a request-projection event; because the offload never reverts, it is one.
**Let each adapter append the replacement.** The adapter owns the budgets but not the session surface; a surface change appended below the loop lets two adapters define the surface differently. Adapters report the count they need instead.
**Plan the offload before dispatch, in the loop or in a plugin.** The loop knows the exact prepared route before deriving each request, so planning there never spends a failed attempt; but it puts a route-specific policy into the one component every profile shares, changes the documented step order, and bypasses the `agent/request-error` waterfall that context-overflow compaction and retry already use for the same repair-and-retry pattern. A pre-step plugin keeps the loop unchanged but cannot see the step's own messages or the first request's route, so the failure path stays necessary, and the plan needs every route to declare its budget on its model info. Handling only the failure costs one attempt per quantum crossing (64 MiB in DeepSeek file mode, 20 MiB on pi-ai) and matches the planned direction: routes will stop checking sizes locally, send every image, and the provider will report what it cannot cache, which is exactly a failure naming an offload point.
**Keep a transient extra offload for the inline fallback and exact-byte overflow.** Would have sent an unlogged projection in exactly the cases the invariant exists for; the failure-and-advance path costs one serialization attempt and keeps every dispatched request derivable from the log.
## Consequences
Offloaded images never return automatically when budgets grow, a larger route is selected, or compaction lowers the total; recovery is the read-only path in the placeholder, which the model uses deliberately. A Files outage or a temporary switch to a small-budget route offloads permanently; both are accepted for the same reason. Each offload copies the carrying node's message into the log; images are references, so the copy is the node's text plus block metadata.
Every dispatched request's image set is determined by the log alone, across file mode, inline fallback, resume, fork, retry, and compaction, and the provider cache prefix no longer oscillates. The execution-world access path embedded in placeholder and handle text is still resolved at serialization time; that gap exists for retained images too and belongs to a separate decision about recording the execution-world mapping. The route-local byte checks are provisional: once the provider reports uncacheable images itself, `requiredImageOffload()` and the route budgets go away while the marked copies and the recovery branch stay as they are.
## Testing
`packages/llm/llm/tests/content.spec.ts` pins arbitrary-depth image traversal, the offload count including the 129-to-64 MiB quantum example, and placeholder projection. `packages/compaction/compaction-image-offload/tests/image-offload.spec.ts` pins the `IMAGE_OFFLOAD_REQUIRED` replace-and-retry path with its `compaction/prune` shadow prices and without a retry event, the untouched original node, request-order counts after an earlier surface replacement, and the exhausted case that delegates downstream. Adapter specs pin placeholder projection, prepared-only-retained reads, and the exact-byte failure with its count; `route-pricing.spec.ts` pins placeholder pricing for a marked replacement. The `inline-image-prompt` TypeScript SDK snapshot replays an authored `IMAGE_OFFLOAD_REQUIRED` attempt through the shipped profile and pins the replacement and the retried request.
@@ -0,0 +1,56 @@
# Agent Note: 以表层替换实现持久的图片 offload
Status: implemented
Archived: 2026-09-10
[English](2026-09-02-durable-image-offload.md) | 中文
## 问题
请求级图片 offload 过去在每次请求时都从头重算。每条路由按最老优先的顺序收集派生表层上的全部图片出现位置,累计字节一旦超过预算,就把超出部分向上取整到整个删除量子,把这么多最老的出现位置替换为占位文本,见[统一图片请求管线](../feature/2026-08-20-unified-image-request-pipeline.zh.md)。没有任何东西记住上一次请求停在哪里,前缀稳定只是因为对 append-only 历史做同样的算术会得到同样的结果。
算术的输入一动,这种稳定就失效。[Files 内联回退](../../archived/bug-fix/2026-08-21-deepseek-files-inline-fallback.md)会用 20 MiB 内联预算和 10 MiB 量子重建请求,那一次省略多得多的图片,下一次 file 模式的请求又把它们带回来。pi-ai 路由的量子是一个字节,前缀几乎每次请求都会移动。compaction 降低总量,让先前省略的图片回归。切换路由会移动每个台阶边界。每一次移动都改变模型可见前缀,并让 provider 的缓存前缀失效。
同样的重算也破坏了仓库不变量:模型可见输入必须能从 session log 重建。实际发出的表示方式、派生请求版本的精确字节长度、路由预算和量子都是运行时或配置事实,从不进入日志,`request/header` 只记录调用配置、系统提示词和工具。provider usage 只锚定 token 总量,恢复不了图片集合,[按路由定价的估计](../../archived/feature/2026-08-24-route-priced-image-request-pressure.md)也写明它不复现回退预算。没有任何消费方能把一条已记录的助手响应和它的请求携带的图片集合配对。
## 决定
一个图片出现位置的省略是一条持久的表层事实,记录方式和 compaction 记录它的缩减一样:一个 `surfaceOp: replace` 节点。
**表层上带标记的副本。** `ImageBlock` 新增 `offloaded?: true`。被省略的出现位置存在于一条与原承载节点同类型的替换事件里(`user/message` 或 `tool/result`),内容是原消息的副本,只把那些块打上标记;`sourceEventSeqs` 指向被替换的节点。session 核心、它的事件表、派生和校验都不变。`deriveMessages()` 原样发送带标记的块;序列化把每个带标记的块渲染为带当前已解析访问路径的 `offloadedImageText`,只准备保留的出现位置。
**只前进。** 替换永不回退。预算变大、路由切换或 compaction 降低总量时,带标记的副本留在原地,所以模型可见前缀和 provider 缓存前缀只向前移动。
**adapter 只投影,不决定。** 支持图片的路由按保留的出现位置的精确请求版本字节执行一个 `LlmImageRequestBudget`(`representation`、`maxBytes`、`maxImages` 与两个量子)。当它们仍超过预算,无论是 file 模式、内联回退更紧的预算还是 pi-ai 上限,adapter 都以 `IMAGE_OFFLOAD_REQUIRED` 让本次尝试失败,并在 `LlmFailure.offloadImages` 中用共享的 `requiredImageOffload()` 算出还需省略多少最老的出现位置。发送前没有任何规划。
**恢复归 `dsh-compaction-image-offload`。** 图片省略是 compaction 在另一个容量维度上的实例:provider 拒绝请求,持久历史被缩减,step 重试。执行器是 compaction 组里 `compaction-tool-result-pruner` 的兄弟包,监听 `agent/request-error` waterfall。收到 `IMAGE_OFFLOAD_REQUIRED` 时,它按模型请求顺序遍历表层,给前 `offloadImages` 个保留的出现位置打标记,为每个承载了其中任一位置的节点先追加 seam 的 `compaction/prune` 影子价格,再追加带标记的副本,然后返回 `retry` 动作,不占提供方重试预算,也不记录 `llm/retry`。assistant 节点承载的是模型输出而不是输入图片,直接跳过。没有可省略的出现位置时向下游委托,失败进入普通恢复路径。循环在替换后的表层上重跑该 step,并像每次表层替换后一样记录新的 `request/header`;agent loop 不变。
**token 记账。** `priceImages` 接收表层的 `ImageBlock`,把带标记的按占位文本定价;DeepSeek 和 replay 的定价不再复现任何 offload 算术。meter 不需要新状态:`compaction/prune` 事件加替换节点,和工具结果剪枝一样重新为该节点定价。
**其他消费方。** compaction 通过 `deriveEventMessage()` 重建每个选中事件,看得到标记。resume、fork 和重放从日志复现表层。纯文本路由保留各自的全历史替换。
## 考虑过的替代方案
**继续每次请求重算 offload 位置。** 只在算术输入不动时稳定,内联回退、pi-ai 量子、compaction 和路由切换都会移动前缀,且没有消费方能重建历史请求的图片集合。
**用 log-only 事件记录每次请求的投影结果。** 恢复了可重建性但没有稳定性:记录的结果不是决策输入,每一种抖动照旧发生,日志只是把它记下来,且省略集合有两个可能不一致的事实来源。
**记录完整的投影后请求体。** 除 offload 决定外一切都已可派生,为记录一个位置而每次请求重复整段历史会让日志平方级增长。
**由 session 派生应用的 log-only `image/offload` 水位事件。** 更早的一版把 offload 点记成一个位置(事件序号加块路径),由 `Session.deriveMessages()` 给位于它及之前的每个出现位置打标记。这让 session 在 `surfaceOp: replace` 之外多出第二种改变模型可见历史的机制,带着自己的校验、折叠、缓存失效和读取时必须识别的事件,并且把逻辑放进了核心而不是做决定的插件。issue #3041 要求永久淘汰使用改变表层的事件而不是请求投影事件;省略永不回退,它就是永久淘汰。
**让各个 adapter 自己追加替换。** adapter 拥有预算,但不拥有会话表层;在循环之下追加表层变更会让两个 adapter 对表层做出不同定义。adapter 改为上报它需要的数量。
**在发送前规划省略,无论放在循环里还是插件里。** 循环在派生每个请求之前就知道精确的已准备路由,在那里规划永远不会多花一次失败的尝试;但这会把一条路由专属的策略放进所有 profile 共用的那个组件,改变已记录的 step 顺序,还绕过了上下文溢出 compaction 和重试已经在用的同一套 `agent/request-error` waterfall。pre-step 插件不改循环,但看不到 step 自己的消息和第一个请求的路由,失败路径仍然必需,而且规划要求每条路由在模型信息上声明预算。只处理失败的代价是每越过一个量子多一次尝试(DeepSeek file 模式 64 MiB,pi-ai 20 MiB),并且和既定方向一致:路由将不再本地检查大小,全部发送,由 provider 报告无法缓存的部分,那正是一个指明省略点的失败。
**为内联回退和精确字节溢出保留临时的额外省略。** 恰好会在不变量所针对的场景发送未记录的投影;失败再推进的路径只多花一次序列化尝试,且让每个已发出请求都可由日志派生。
## 后果
预算变大、选中更大的路由或 compaction 降低总量时,被省略的图片不会自动回归;恢复手段是占位文本中的只读路径,模型需要时主动使用。一次 Files 故障或临时切到小预算路由会永久省略,两者出于同一理由被接受。每次省略会把承载节点的消息复制一份进日志;图片只是引用,复制的是该节点的文本和块元数据。
每个已发出请求的图片集合都仅由日志决定,覆盖 file 模式、内联回退、resume、fork、重试和 compaction,provider 缓存前缀不再来回变化。占位文本和句柄文本里嵌入的执行世界访问路径仍在序列化时解析;这个缺口对保留的图片同样存在,属于另一个关于记录执行世界映射的决定。路由本地的字节检查是过渡实现:provider 自己报告无法缓存的图片之后,`requiredImageOffload()` 和路由预算会被删掉,带标记的副本和恢复分支保持不变。
## 测试
`packages/llm/llm/tests/content.spec.ts` 钉住任意深度的图片遍历、包括 129 到 64 MiB 量子示例在内的省略计数,以及占位投影。`packages/compaction/compaction-image-offload/tests/image-offload.spec.ts` 钉住 `IMAGE_OFFLOAD_REQUIRED` 替换并重试、带 `compaction/prune` 影子价格且不追加重试事件的路径、原节点保持不变、更早一次表层替换之后的请求顺序计数,以及向下游委托的耗尽情况。adapter 测试钉住占位投影、只读取保留图片以及带数量的精确字节失败;`route-pricing.spec.ts` 钉住带标记替换的占位定价。`inline-image-prompt` TypeScript SDK 快照通过发布的 profile 重放一次手工编写的 `IMAGE_OFFLOAD_REQUIRED` 尝试,钉住替换和重试后的请求。
@@ -1,6 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/process/2026-09-08-comment-only-review-routing.md
2026-09-08-comment-only-review-routing.md: 050905285b2291b34da9873d19c2f122c088a9e5
2026-09-08-comment-only-review-routing.zh.md: b98f5d70b4d5c0fd27df1c393238b0802e420e09
# pnpm run verify-translation-pairing --write .agents/notes/implemented/feature/2026-09-08-web-explicit-file-delivery.md
2026-09-08-web-explicit-file-delivery.md: ff2ddeb59ded05b70006dce217df2966eab9d4f2
2026-09-08-web-explicit-file-delivery.zh.md: 85b09ac82c83365b1c198168b78aa7ac84ef216f
@@ -0,0 +1,38 @@
# Agent Note: Web delivers explicit file snapshots
Status: implemented
Archived: 2026-09-08
English | [中文](2026-09-08-web-explicit-file-delivery.zh.md)
## Problem
Workspace links read live paths, so edits or deletion can invalidate a final deliverable. Files created through shell commands also lack first-party editor mutation records. Delivery needs an explicit operation and saved bytes without expanding Session ZIP exports.
## Decision
The [present tool](../../../../packages/fs/tool-present/README.md) owns execution, immutable snapshots, delivery types, and the durable event. The [deliverables plugin](../../../../packages/client/ui-deliverables/README.md) owns authenticated snapshot actions and browser rendering, with type-only imports from the tool’s `./types` entry. The `standard`, `ptc`, and `cordis` presets mount the tool package; `minimal` retains its two-tool training configuration. The existing attachment service saves immutable bytes; successful final `tools/result` notifications append `deliverables/presented` to the calling Session. Native and nested calls use the same recorder. A later enclosing program failure does not undo a completed nested delivery. Blocked tool results publish none.
Download and native-open requests authorize a reference by the viewed Session, event sequence, and file index. The event stores no Session ID, so forked history uses the child's own log. The existing produced-file row keeps its names and behavior. Session ZIP retains delivery events but does not collect their attachment bytes.
Card and closing-mention gestures open a verified private copy with the existing native-command utility. A POST expresses the desktop side effect; GET remains a byte read. Each gesture receives a new copy so application edits cannot corrupt the immutable attachment or alter later opens. Successful copies survive until plugin disposal for applications that read lazily; failed copies are removed immediately, and disposal awaits cancelled work before cleanup.
## Alternatives considered
**A Host tool subpath in the UI package** couples preset installation to browser packaging and requires extra published entries. An ordinary tool package preserves shared filesystem and tool error classes through the repository’s peer dependency rules.
**Live workspace links** cannot preserve a delivered version after edits or deletion. Opening the attachment store’s own path instead would expose immutable saved bytes to application writes.
**Generic artifact fields throughout tools, dispatch, and Session** would broaden unrelated APIs for one Web feature. A plugin-owned event uses existing extension points and avoids parent-result forwarding.
**Tool text as the durable index** is unreliable because post-processing and spill can replace ordinary or nested result text. Each plugin instance retains its own completed snapshots by execution identity and publishes them only on a successful final result. Same-name scoped replacements cannot create or duplicate another instance’s delivery records.
**Descriptor-bound filesystem extensions** would change multiple capability providers. This feature uses existing bounded reads with containment and before/after version checks. Those checks reject ordinary concurrent changes but do not guarantee atomic confinement against swap-and-restore; stronger filesystem guarantees belong to the filesystem provider.
## Consequences
The implementation adds no artifact service or attachment format. Unreferenced snapshots can remain after partial failure; attachment retention remains service-owned. A downstream build must understand the new required event to read the log. The generated Session event inventory records that requirement without changing released format generations.
The delivery event is required-on-read because it is the authorization index for saved bytes, not only display metadata. Skipping it would allow an older reader to reconstruct or fork a Session without its completed deliveries. Unsupported readers refuse that loss instead of silently dropping the references.
Focused tests cover snapshot bytes, invalid inputs, blocked results, HTTP integrity, native-open copy isolation, retry and disposal, turn isolation, and fork-addressed actions. The recorded Web scenario covers nested completion followed by an enclosing failure, source deletion, reload, native-open gestures without browser downloads, and ZIP exclusion.
@@ -0,0 +1,38 @@
# Agent Note: Web 显式交付文件快照
Status: implemented
Archived: 2026-09-08
[English](2026-09-08-web-explicit-file-delivery.md) | 中文
## 问题
工作区链接读取当前路径,因此编辑或删除会使最终交付文件失效。通过 shell 命令创建的文件也没有第一方编辑器修改记录。交付需要显式操作和保存的字节,同时不扩大 Session ZIP 导出内容。
## 决策
[present 工具](../../../../packages/fs/tool-present/README.zh.md)拥有执行、不可变快照、交付类型和持久事件。[交付插件](../../../../packages/client/ui-deliverables/README.zh.md)拥有认证快照操作和浏览器渲染,仅从工具的 `./types` 入口导入类型。`standard`、`ptc` 与 `cordis` preset 挂载工具包;`minimal` 保留双工具训练配置。现有 attachment 服务保存不可变字节;成功的最终 `tools/result` 通知将 `deliverables/presented` 追加到调用方 Session。原生与嵌套调用使用同一个记录器。外层程序随后失败不会撤销已完成的嵌套交付。被阻止的工具结果不发布交付。
下载与原生打开请求通过当前查看的 Session、事件序号与文件索引授权引用。事件不保存 Session ID,因此 fork 历史使用子 Session 自己的日志。现有产出文件行保留其名称和行为。Session ZIP 保留交付事件,但不收集其中引用的 attachment 字节。
卡片和收尾引用操作通过现有 native-command 工具,在默认应用中打开经过校验的私有副本。POST 表达桌面副作用;GET 仍仅读取字节。每次操作创建新副本,避免应用内编辑损坏不可变 attachment 或改变后续打开的内容。成功副本保留到插件释放,以支持延迟读取的应用;失败副本立即删除,释放时先等待取消的操作结束再清理。
## 已考虑的替代方案
**在 UI 包中提供 Host 工具子路径**会将 preset 安装与浏览器打包耦合,并要求额外发布入口。普通工具包通过仓库 peer dependency 规则保留共享的文件系统和工具错误类。
**实时工作区链接**无法在编辑或删除后保留已交付版本。直接打开 attachment 存储路径则会使不可变保存字节暴露于应用写入。
**在工具、dispatch 和 Session 中增加通用 artifact 字段**会为单个 Web 功能扩大无关 API。插件拥有的事件使用现有扩展点,并省去父调用结果转发。
**将工具文本作为持久索引**并不可靠,因为后处理与 spill 可以替换普通或嵌套结果文本。每个插件实例按执行对象保留自身已完成的快照,仅在最终结果成功时发布。同名作用域替代工具不能创建或重复其他实例的交付记录。
**基于文件描述符的文件系统扩展**会修改多个能力提供方。本功能使用现有有界读取,并检查路径包含关系及读取前后的版本。这些校验会拒绝普通并发变化,但不保证对替换后复原提供原子路径限制;更强的文件系统保证属于文件系统提供方。
## 影响
实现不增加 artifact 服务或 attachment 格式。部分失败后可能留下无引用快照;attachment 保留策略仍由服务拥有。下游构建必须理解新必需事件才能读取日志。生成的 Session 事件清单记录该要求,不修改已发布的格式代际。
交付事件要求读取端识别,因为它是保存字节的授权索引,不只是显示元数据。跳过事件会让旧读取端在重建或分叉 Session 时丢失已完成的交付。不支持该事件的读取端拒绝读取,避免静默丢弃引用。
定向测试覆盖快照字节、无效输入、被阻止的结果、HTTP 完整性、原生打开的副本隔离、重试与释放、turn 隔离及使用 fork 地址的操作。录制 Web 场景覆盖嵌套调用完成后外层失败、源文件删除、重新加载、不触发浏览器下载的原生打开操作和 ZIP 排除。
+6
View File
@@ -358,6 +358,9 @@
"architecture/2026-09-01-streamed-tool-call-identity.i18n.yaml": "sha256:5193bf7429a0300f743ea7e6087c3175a7d28c2c468ca1cfd1f77702b68b6399",
"architecture/2026-09-01-streamed-tool-call-identity.md": "sha256:303433b95c22571bf6a801da604b42a4c7231f5f6d07daeb518354f0d7019141",
"architecture/2026-09-01-streamed-tool-call-identity.zh.md": "sha256:53728d3ee06b29d6023e28024fb3ab6fa4a7473ae5e55368cd19690719d120f3",
"architecture/2026-09-02-durable-image-offload.i18n.yaml": "sha256:d06155597e7df4ca4c15a2902ab664d69cc2fde1e7963e61543d20a49930d2b9",
"architecture/2026-09-02-durable-image-offload.md": "sha256:02b7becba3dd73a4e627b85705e8bd5dd55978cb041f4d3a0fd1420a76e4de92",
"architecture/2026-09-02-durable-image-offload.zh.md": "sha256:a0464b8272086d663ee698efd0b2edd6224d40770d7a8fd88a6d8bfe030b2a1d",
"architecture/2026-09-02-protocol-specific-model-listing-discovery.i18n.yaml": "sha256:ba8a62a9fa3263de4162709cb1aaf81f6ad4feee5011d7a5d8b24eb92ad81d30",
"architecture/2026-09-02-protocol-specific-model-listing-discovery.md": "sha256:9db9c47559c17f0e2931b9e646f1d2d17b590df4d5699d4e779ed6d7f65fc371",
"architecture/2026-09-02-protocol-specific-model-listing-discovery.zh.md": "sha256:9a266590d49093a097bcdb5d09865757cdd07625cfaaddce1303046d580551f1",
@@ -1381,6 +1384,9 @@
"feature/2026-09-01-web-superellipse-corner-smoothing.i18n.yaml": "sha256:50afdbe5b5e19889918af6d86ab3218c05205be35938b6d33d158c60777e3b58",
"feature/2026-09-01-web-superellipse-corner-smoothing.md": "sha256:b1445101c49e74bbcb4f607af850cd6df105d4034828d0dd47081e8079148f15",
"feature/2026-09-01-web-superellipse-corner-smoothing.zh.md": "sha256:1a278c417c0d7de3b4c3c35061b419303b4a1a0707831c283d8f862ab9b6fd23",
"feature/2026-09-08-web-explicit-file-delivery.i18n.yaml": "sha256:99daae539cc8fd7376ce0265538bee21e1e33f3c0d77c8cc4e011f94b4e9568a",
"feature/2026-09-08-web-explicit-file-delivery.md": "sha256:bb416b1e8be081e6cb6af17792eb1a442172ff114c3a3577e5ef06a77eb57093",
"feature/2026-09-08-web-explicit-file-delivery.zh.md": "sha256:00a642380e1f6ac9d5cd840e021f4e3e4ae68a289cb6344eb3dcd73a6fd81f4d",
"process/2026-06-11-doc-sync-enforcement.i18n.yaml": "sha256:33b6d5874427bd7a2bd82e7e2f4f482b12448b2464aef15a9c57975edb48554d",
"process/2026-06-11-doc-sync-enforcement.md": "sha256:aa2fe83d519fc30d48dff19e596e83c8922aacc9e063e14fe2cc35b769b9100e",
"process/2026-06-11-doc-sync-enforcement.zh.md": "sha256:698017bd35f030fdea3eac51df9e43138c48140f504739d687b7251d13fced2b",
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-11-dev-invariants-over-deep-readonly.md
2026-06-11-dev-invariants-over-deep-readonly.md: 9adb741db1e26e3a9f750fed60524351159cfeaa
2026-06-11-dev-invariants-over-deep-readonly.zh.md: d352eca985b6594ec5d10c9b73c2ecc18ea458b5
2026-06-11-dev-invariants-over-deep-readonly.md: 0784960397f4d198e136e284ef9232b4ed3e2c2c
2026-06-11-dev-invariants-over-deep-readonly.zh.md: 7183813130560bf82426c8719570b668ba86b7e5
@@ -22,7 +22,7 @@ Responsibility is split between an always-on storage boundary and optional devel
`Session` accepts an event only after one recursive pass has materialized a lossless JSON snapshot. That pass rejects unsupported values and produces the exact detached record that enters the log, so validation and storage cannot observe different values from a stateful getter or retain caller-owned nested references.
The accepted event and all of its descendants are deep-frozen before publication. `append()` returns that owned frozen event, and `session/event` observers and `eventAt(seq)` receive the same record. `snapshotEvents(fromSeq?, toSeqExclusive?)` returns a frozen array snapshot; a previously returned array does not grow after a later append. `seq` and `eventAt()` avoid array materialization when a caller needs only the current length or one event. Seed records pass through the same validation, snapshot, and freeze boundary before construction succeeds.
The accepted event and all of its descendants are deep-frozen before publication. `append()` returns that owned frozen event, and `session/event` observers and `eventAt(seq)` receive the same record. `snapshotEvents(fromSeq?, toSeqExclusive?)` returns a frozen array snapshot; a previously returned array does not grow after a later append. `seq` reads the current length without materializing an array. Synchronous historical readers are deprecated under the [event-read policy](2026-09-09-deprecate-synchronous-session-event-reads.md). Seed records pass through the same validation, snapshot, and freeze boundary before construction succeeds.
This guarantee belongs in `Session`, not in an optional listener, because every composition relies on trustworthy history. A production deployment, a focused test, or a custom embedding receives the same storage semantics whether or not development support plugins are registered.
@@ -53,7 +53,7 @@ Detaching `deriveMessages()` would protect the most common request path but leav
## Consequences
- Every accepted live or seeded session event is detached from caller-owned inputs and deeply immutable before any observer can receive it.
- `snapshotEvents()` exposes stable immutable snapshots instead of the private growing array; `seq` and `eventAt()` serve scalar reads without copying that array.
- Existing `snapshotEvents()` and `eventAt()` callers retain immutable read results while their migration is deferred; `seq` reads the log length without copying the array.
- Request-side mutation cannot reach stored history through derived messages.
- Development builds can enable relational assertions without changing storage behavior, and disposing or filtering a companion does not weaken log immutability.
- `dsh-invariants` configures global enablement plus package allow/block regex lists; each check remains owned and tested by its product package.
@@ -22,7 +22,7 @@ TypeScript readonly 类型不是充分的运行时边界。它们在程序运行
`Session` 仅在一次递归遍历完成无损 JSON 快照的物化之后才接受事件。该遍历拒绝不支持的值,并产出进入日志的已分离的确切记录,因此验证与存储不会从有状态的 getter 观察到不同的值,也不会保留调用方拥有的嵌套引用。
被接受的事件及其所有后代在发布前被深度冻结。`append()` 返回由 Session 拥有的冻结事件,`session/event` 观察者和 `eventAt(seq)` 接收同一记录。`snapshotEvents(fromSeq?, toSeqExclusive?)` 返回冻结的数组快照;先前返回的数组不会因后续 append 而增长。调用方只需要当前长度或单个事件时,`seq` 和 `eventAt()` 不会物化数组。种子记录在构造成功前经过相同的验证、快照与冻结边界。
被接受的事件及其所有后代在发布前被深度冻结。`append()` 返回由 Session 拥有的冻结事件,`session/event` 观察者和 `eventAt(seq)` 接收同一记录。`snapshotEvents(fromSeq?, toSeqExclusive?)` 返回冻结的数组快照;先前返回的数组不会因后续 append 而增长。`seq` 无需物化数组即可读取当前长度。同步历史读取方法按[事件读取策略](2026-09-09-deprecate-synchronous-session-event-reads.zh.md)弃用。种子记录在构造成功前经过相同的验证、快照与冻结边界。
此保证属于 `Session` 而非可选监听器,因为每种组合都依赖可信的历史。无论是否注册了开发支持插件,生产部署、聚焦测试或自定义嵌入都获得相同的存储语义。
@@ -53,7 +53,7 @@ TypeScript readonly 类型不是充分的运行时边界。它们在程序运行
## 后果
- 每个被接受的实时或种子会话事件在任何观察者接收之前,都已从调用方拥有的输入中分离并深度不可变。
- `snapshotEvents()` 暴露稳定的不可变快照,而非持续增长的私有数组;`seq` 和 `eventAt()` 为标量读取提供无需复制数组的路径。
- 现有 `snapshotEvents()` 和 `eventAt()` 调用方在暂缓迁移期间仍获得不可变的读取结果;`seq` 无需复制数组即可读取日志长度。
- 请求侧的修改无法通过派生消息触及已存储的历史。
- 开发构建可以启用关系断言而不改变存储行为;dispose 或过滤一个配套插件不会削弱日志不可变性。
- `dsh-invariants` 配置全局启用状态以及包名允许/阻止 regex 列表;每项检查仍由其产品包拥有并测试。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-13-capability-seams.md
2026-06-13-capability-seams.md: 3c552c474b499b9f1c9f60242f4773750faadafb
2026-06-13-capability-seams.zh.md: 4e6100a2ce557d91fa18b5630c267f37f1bc0f00
2026-06-13-capability-seams.md: 798739b7ef203b10ea657f9e1c178ea7eab094ad
2026-06-13-capability-seams.zh.md: c8f568aa569fe74146f1e7fcc46e00774a39fb84
@@ -8,7 +8,7 @@ English | [中文](2026-06-13-capability-seams.zh.md)
The harness has swappable capabilities, including shell execution and model providers. A capability has three concerns that change at different rates and for different reasons: the *contract* (what the capability is), the *implementation* (how it runs), and the *consumer API* (what the model and other plugins program against). Bundling them in one package couples those rates of change — swapping a local executor for a sandboxed one would churn the tool schemas the model sees, even though the model-facing contract never changed.
This is distinct from "who provides vs. needs a capability at runtime", which Cordis already answers with services + `inject` (a provider registers `ctx.shell`; a consumer declares `inject: ['bash']` and its fiber pends until the service exists). That mechanism is necessary but doesn't dictate package boundaries; this Agent Note does.
This is distinct from "who provides vs. needs a capability at runtime", which Cordis already answers with services + `inject` (a provider registers `ctx.shell`; a consumer declares `inject: ['shell']` and its fiber pends until the service exists). That mechanism is necessary but doesn't dictate package boundaries; this Agent Note does.
## Decision
@@ -8,7 +8,7 @@ Status: implemented
harness 具有可替换的能力,包括 shell 执行和模型提供方。一项能力涉及三个关注点,它们以不同速率、因不同原因变化:*约定*(这项能力是什么)、*实现*(它如何运行)、*消费方 API*(模型和其他插件面向什么编程)。将三者捆绑在一个包中会耦合这些变化速率——把本地执行器换成沙箱化执行器时,模型看到的工具 schema 也会被搅动,尽管面向模型的约定从未改变。
这与「谁在运行时提供、谁需要一项能力」是不同的问题,后者 Cordis 已通过服务 + `inject` 解决(提供方注册 `ctx.shell`;消费方声明 `inject: ['bash']`,其 fiber 挂起直到服务存在)。该机制是必要的,但不决定包的边界;本 Agent Note 决定的是包的边界。
这与「谁在运行时提供、谁需要一项能力」是不同的问题,后者 Cordis 已通过服务 + `inject` 解决(提供方注册 `ctx.shell`;消费方声明 `inject: ['shell']`,其 fiber 挂起直到服务存在)。该机制是必要的,但不决定包的边界;本 Agent Note 决定的是包的边界。
## 决策
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-13-twin-llm-adapters.md
2026-06-13-twin-llm-adapters.md: a4c87325a0b0d1ebe6cf8f95672e5de74ef37d57
2026-06-13-twin-llm-adapters.zh.md: 36996750a16cc95393d727cffee3bcc53573eaf2
2026-06-13-twin-llm-adapters.md: fe8b0b55e0e027e29eb0920e64a1760d0dc35aee
2026-06-13-twin-llm-adapters.zh.md: 4248b5afeb9f3fd4503914e53203c674e3a80dbf
@@ -25,3 +25,5 @@ The rule they enforce: **anything the StreamChunk vocabulary cannot express for
## Consequences
The twin doubles adapter and key-gated e2e maintenance—both cover V4 Flash and Pro across representative reasoning modes—in exchange for continuous seam-neutrality validation and a second implementation example. Both use `apiKey`, `baseURL`, and `models`; the direct-fetch adapter exposes `thinking`/`reasoningEffort`, while pi-ai exposes one `reasoning` level. A future conformance suite could justify retiring one adapter through a superseding Agent Note.
The [Messages adapter](../feature/2026-09-07-deepseek-messages-adapter.md) adds an Anthropic-protocol implementation inside `llm-deepseek`; it preserves the same stream conventions.
@@ -25,3 +25,5 @@ Status: implemented
## 后果
孪生体使适配器和需要密钥的 e2e 维护量翻倍——两者都覆盖 V4 Flash 和 Pro 在各代表性推理(reasoning)模式下的行为——换来的是持续的 seam 中立性验证和第二份实现示例。两个适配器均使用 `apiKey`、`baseURL` 和 `models`;直接 fetch 适配器暴露 `thinking`/`reasoningEffort`,pi-ai 适配器暴露一个 `reasoning` 级别。未来如果有一致性测试套件,可以通过后续 Agent Note 论证退役其中一个适配器。
[Messages 适配器](../feature/2026-09-07-deepseek-messages-adapter.zh.md) 在 `llm-deepseek` 内增加 Anthropic 协议实现,并遵守相同的流约定。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-14-session-persistence.md
2026-06-14-session-persistence.md: 55c1bbab94bb7854fd6bbcaace56c30dbcdb01dc
2026-06-14-session-persistence.zh.md: 16546ab61773da47064de8388803e92c8b454eae
2026-06-14-session-persistence.md: d79975e1fedb6efcd0e4ea83bc799bb158082e41
2026-06-14-session-persistence.zh.md: f70b8005a254cc6a9ea8ada31c32e5552ee0c8b2
@@ -15,11 +15,11 @@ The [event-sourced model](2026-06-11-event-sourced-sessions.md) makes the append
Persistence is a **capability seam** with an abstract Service Definition ([capability seams](2026-06-13-capability-seams.md), the `dsh-shell` template), not loop or core logic:
1. **Interface** (`dsh-session-persistence`, `ctx.sessionPersistence`) — an abstract `SessionPersistence` service: `create`/`open`/`stat`/`list`/`flush`, with `create`/`open` returning per-session `SessionHandle`s that carry `read`/`append`/`flush`/`close` ([handle-based seam](2026-08-27-handle-based-session-persistence.md)). Its persisted unit IS the existing `SessionEvent` (`{ type, seq, time, data }`), reused verbatim — no conversion type.
2. **Implementation** (`dsh-session-persistence-jsonl`) — an append-only logical JSONL log per session: a `SessionHeader` line followed by storage records that losslessly represent the contiguous `SessionEvent` stream. Current v2 writes one event per row; frozen v0 and v1 readers retain their historical packed-delta representation. [Checksummed Zstandard frames](2026-07-19-zstandard-jsonl-session-logs.md) are the default physical encoding, with raw lines configurable.
2. **Implementation** (`dsh-session-persistence-jsonl`) — an append-only logical JSONL log per session: a `SessionHeader` line followed by storage records that losslessly represent the contiguous `SessionEvent` stream. The current format writes one event per row; frozen v0 and v1 readers retain their historical packed-delta representation. [Checksummed Zstandard frames](2026-07-19-zstandard-jsonl-session-logs.md) are the default physical encoding, with raw lines configurable.
Key durable, contested choices:
- **The canonical durable log persists every current `SessionEvent` losslessly.** In v2, one `assistant/message` or `assistant/attempt` embeds the exact timed provider stream for an attempt; `deriveMessages()` projects only the surface message. Dropping embedded stream members is tempting, but it loses replay, timing, usage, partial-failure, and diagnostic facts. Removing a complete event likewise requires dense renumbering because `seq = log.length` and `events[i].seq === i`; the [v1-to-v2 migration](2026-09-01-v2-embedded-assistant-streams.md) performs that rewrite explicitly rather than filtering the canonical log.
- **The canonical durable log persists every current `SessionEvent` losslessly.** One `assistant/message` or `assistant/attempt` embeds the exact timed provider stream for an attempt; `deriveMessages()` projects only the surface message. Dropping embedded stream members is tempting, but it loses replay, timing, usage, partial-failure, and diagnostic facts. Removing a complete event likewise requires dense renumbering because `seq = log.length` and `events[i].seq === i`; the [v1-to-v2 migration](2026-09-01-v2-embedded-assistant-streams.md) performs that rewrite explicitly rather than filtering the canonical log.
- **Append-only; a crashed turn is closed, never truncated.** Flushed events are never rewritten. The [semantic checkpoint policy](../../../../packages/session/session-checkpoint-policy/README.md) drains the request before model dispatch, a recorded top-level call before tool dispatch, and the complete response/result batch after a step; the loop drains the final turn boundary. Because one interrupted turn may contain substantial valid work, persistence returns its contiguous, parseable events unmodified; the reader owns balancing — resume computes risk-classified error results for unanswered assistant calls, a missing `step/end`, and `turn/end` with `{ kind: 'interrupted' }` (`interruptedTurnClosers`) and appends them through its write handle, while read-only observers add the same closers in memory. The synthetic results keep resumed provider transcripts valid. Only the incomplete fragment of a torn final append is discarded — complete records recovered from it are durably rewritten by the write path before its first new append; a parse error or sequence gap in the committed prefix is corruption and makes the session unloadable.
- **The file backend is canonical while the service remains extensible.** `dsh-session-persistence-jsonl` is the sole first-party provider and passes `runPersistenceContract`; the abstract service remains available to out-of-tree providers. The [JSONL-only persistence decision](../simplification/2026-08-30-jsonl-only-session-persistence.md) owns removal of the first-party database provider and its deliberate compatibility cut.
- **Metadata is out-of-log.** Format version, cwd, and lineage are storage concerns, not replayable conversation state, so they live in a `SessionHeader` owned by `dsh-session` and attached to a `Session` via a new readonly `session.header` — never in `SessionEventMap`, never reaching `deriveMessages()`. `createdAt` is non-negative safe-integer Unix epoch milliseconds: live creation and persistence registration reject fractional values, and JSONL validates the decoded header. The alternative (a merge-extensible `session/meta` event as log line 0) was rejected: an in-log event would ride along with a seeded/forked session for free, but metadata is not replayable state, so the explicit out-of-log header boundary is the cleaner cost. (The header was originally split into an immutable `SessionHeader` plus a mutable `SessionSummary` whose union was `SessionMeta`; the mutable summary was later removed as dead state — see [Drop the mutable session summary](../../archived/simplification/2026-06-19-drop-mutable-session-summary.md).)
@@ -29,7 +29,7 @@ Key durable, contested choices:
Each key choice above records its rejected alternative where the choice is stated: a **stream-filtered canonical log** — loses attempt evidence, while removing events without an explicit migration breaks contiguous sequence numbers; **truncating a crashed turn** — silently destroys a long autonomous run's real work; an **in-log `session/meta` event as log line 0** — metadata is not replayable state; **finite fractional `createdAt` values** — have no producer and diverge from integer Unix-millisecond storage; **hard-injecting `sessionPersistence` into the loop** — would pend non-persistent demos forever.
Format versioning: the header carries a `version`; handles expose only `SESSION_FORMAT_VERSION = 2`. JSONL event-body reads compose the static v0-to-v1 and v1-to-v2 adjacent migration chain before returning a handle; the first edge owns bounded legacy normalization, while the second owns Assistant stream embedding and dense reference remapping. V0 remains at suffixless `session.jsonl[.zstd]`, while positive versions use immutable lowercase `session.vN.jsonl[.zstd]` names ([released Session migration](2026-08-31-released-session-format-migrations.md)). Current-generation append and flush are robust to partial trailing writes; a future provider or write-ahead log needs its own power-loss and recovery contract.
Format versioning: the header carries a `version`; handles expose only the logical format selected by `SESSION_FORMAT_VERSION` ([version authority](../../../../docs/session-format-status.md)). JSONL event-body reads compose the complete static adjacent migration chain before returning a handle; each edge owns its historical transformations. V0 remains at suffixless `session.jsonl[.zstd]`, while positive versions use immutable lowercase `session.vN.jsonl[.zstd]` names ([released Session migration](2026-08-31-released-session-format-migrations.md)). Current-generation append and flush are robust to partial trailing writes; a future provider or write-ahead log needs its own power-loss and recovery contract.
## Consequences
@@ -15,11 +15,11 @@ Status: implemented
持久化是一个具有抽象 Service Definition 的**能力 seam**([能力 seam](2026-06-13-capability-seams.zh.md),`dsh-shell` 模板),而非循环或核心逻辑:
1. **接口**(`dsh-session-persistence`,`ctx.sessionPersistence`):一个抽象的 `SessionPersistence` 服务,提供 `create`/`open`/`stat`/`list`/`flush`,其中 `create`/`open` 返回逐会话的 `SessionHandle`,句柄承载 `read`/`append`/`flush`/`close`([基于句柄的 seam](2026-08-27-handle-based-session-persistence.zh.md))。其持久化单元就是现有的 `SessionEvent`(`{ type, seq, time, data }`),原样复用,无转换类型。
2. **实现**(`dsh-session-persistence-jsonl`):每个会话一个仅追加的逻辑 JSONL 日志:先是一行 `SessionHeader`,随后是无损表示连续 `SessionEvent` 流的存储记录。当前 v2 每个事件写一行;冻结的 v0 与 v1 reader 保留其历史 packed-delta 表示。[带校验和的 Zstandard 帧](2026-07-19-zstandard-jsonl-session-logs.zh.md)是默认物理编码,也可通过配置使用原始行。
2. **实现**(`dsh-session-persistence-jsonl`):每个会话一个仅追加的逻辑 JSONL 日志:先是一行 `SessionHeader`,随后是无损表示连续 `SessionEvent` 流的存储记录。当前格式每个事件写一行;冻结的 v0 与 v1 reader 保留其历史 packed-delta 表示。[带校验和的 Zstandard 帧](2026-07-19-zstandard-jsonl-session-logs.zh.md)是默认物理编码,也可通过配置使用原始行。
长期有效、存在争议的关键选择:
- **规范持久日志无损保留每个当前 `SessionEvent`。** 在 v2 中,一个 `assistant/message` 或 `assistant/attempt` 会嵌入该 attempt 的精确带时间 provider stream;`deriveMessages()` 只投影 surface message。删除嵌入 stream 成员看似诱人,但会丢失 replay、timing、usage、部分失败与诊断事实。移除完整事件同样需要密集重新编号,因为 `seq = log.length` 且 `events[i].seq === i`;[v1 到 v2 迁移](2026-09-01-v2-embedded-assistant-streams.zh.md)会显式执行该改写,而不是过滤规范日志。
- **规范持久日志无损保留每个当前 `SessionEvent`。** 一个 `assistant/message` 或 `assistant/attempt` 会嵌入该 attempt 的精确带时间 provider stream;`deriveMessages()` 只投影 surface message。删除嵌入 stream 成员看似诱人,但会丢失 replay、timing、usage、部分失败与诊断事实。移除完整事件同样需要密集重新编号,因为 `seq = log.length` 且 `events[i].seq === i`;[v1 到 v2 迁移](2026-09-01-v2-embedded-assistant-streams.zh.md)会显式执行该改写,而不是过滤规范日志。
- **仅追加;崩溃的轮次被关闭,而非截断。** 已刷写的事件永不被重写。[语义检查点策略](../../../../packages/session/session-checkpoint-policy/README.zh.md)会在调用模型前排空请求、在调用工具前排空已记录的顶层调用,并在步骤结束后排空完整的响应/结果批次;循环则排空最终轮次边界。由于一个被中断的轮次可能包含大量有效工作,持久化会原样返回其连续、可解析的事件;配平是读方的职责——resume 会为未应答的 assistant 调用计算按风险分类的错误结果、补一个缺失的 `step/end`,以及带 `{ kind: 'interrupted' }` 的 `turn/end`(`interruptedTurnClosers`),并通过其写句柄追加它们,而只读观察方仅在内存中添加同样的收尾事件。合成结果保证恢复后的提供方 transcript(文本记录)仍然有效。只有撕裂的最终 append 中不完整的碎片会被丢弃——从中恢复的完整记录由写路径在第一次新 append 之前持久重写;已提交前缀中的解析错误或序号间隙,属于数据损坏,会使该会话不可加载。
- **文件后端为规范实现,服务保持可扩展。** `dsh-session-persistence-jsonl` 是唯一 first-party provider,并通过 `runPersistenceContract`;抽象服务继续供仓库外 provider 使用。[JSONL-only 持久化决策](../simplification/2026-08-30-jsonl-only-session-persistence.zh.md)负责 first-party 数据库 provider 的删除及其明确 compatibility cut。
- **元数据在日志之外。** 格式版本、cwd 和谱系是存储关注点,不是可回放的对话状态,因此它们存放在 `dsh-session` 拥有的 `SessionHeader` 中,并通过新的只读属性 `session.header` 附加到 `Session` 上——永远不进入 `SessionEventMap`,永远不到达 `deriveMessages()`。`createdAt` 是以 Unix epoch 毫秒表示的非负安全整数:运行时创建和持久化注册会拒绝小数值,JSONL 会验证解码后的 header。替代方案(一个可合并扩展的 `session/meta` 事件作为日志第 0 行)被否决:日志内事件会自然随 seed/fork 的会话携带,但元数据不是可回放状态,因此显式的日志外 header 边界是更清晰的取舍。(header 最初被拆分为不可变的 `SessionHeader` 加可变的 `SessionSummary`,二者的联合类型为 `SessionMeta`;可变 summary 后来因属于死状态而被移除——见 [移除可变会话摘要](../../archived/simplification/2026-06-19-drop-mutable-session-summary.md)。)
@@ -29,7 +29,7 @@ Status: implemented
上述每个关键选择都在陈述处记录了被否决的替代方案:**过滤 stream 的规范日志**会丢失 attempt 证据,而未通过显式迁移移除事件会破坏连续序号;**截断崩溃的轮次**会静默销毁长时间自主运行中的真实工作;**日志内 `session/meta` 事件作为第 0 行**——元数据不是可回放状态;**有限的非整数 `createdAt` 值**没有生产方,且与整数 Unix 毫秒存储不一致;**将 `sessionPersistence` 硬注入循环**会让非持久化的演示永远挂起。
格式版本控制:header 携带 `version`;句柄只暴露 `SESSION_FORMAT_VERSION = 2`。JSONL 的事件正文读取会在返回句柄前组合静态 v0-to-v1 与 v1-to-v2 相邻迁移链;第一条边负责有界 legacy normalization,第二条边负责 Assistant stream 嵌入与密集引用重映射。V0 保留无后缀的 `session.jsonl[.zstd]`,正版本则使用不可变的小写 `session.vN.jsonl[.zstd]` 名称([已发布 Session 迁移](2026-08-31-released-session-format-migrations.zh.md))。当前 generation 的 append 与 flush 能稳健处理不完整尾部写入;未来 provider 或 WAL 必须定义自己的断电与恢复约定。
格式版本控制:header 携带 `version`;句柄只暴露由 `SESSION_FORMAT_VERSION` 选定的逻辑格式([版本真源](../../../../docs/session-format-status.zh.md))。JSONL 的事件正文读取会在返回句柄前组合完整的静态相邻迁移链;每条迁移边拥有自身的历史转换。V0 保留无后缀的 `session.jsonl[.zstd]`,正版本则使用不可变的小写 `session.vN.jsonl[.zstd]` 名称([已发布 Session 迁移](2026-08-31-released-session-format-migrations.zh.md))。当前 generation 的 append 与 flush 能稳健处理不完整尾部写入;未来 provider 或 WAL 必须定义自己的断电与恢复约定。
## 后果
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-06-18-session-surface.md
2026-06-18-session-surface.md: 93ea55883dedd943fe1ffac67a9842c962ca6dac
2026-06-18-session-surface.zh.md: 54ecb1162bc46007dfcbb7d8cb39075d52171567
2026-06-18-session-surface.md: 26780d54a1ea12283900572c6fd0e8f740de5524
2026-06-18-session-surface.zh.md: 328a9ee9f1bed9c52996cfd307549a7745e16d4b
@@ -37,7 +37,7 @@ Delta processing is O(1) when no new events and O(new events) when new events ar
### Persistence
The fields are serialized as top-level JSON properties. JSONL preserves placement and provenance without a separate column mapping. The [V3 canonical-envelope decision](2026-09-06-v3-canonical-session-envelopes.md) owns exact replacement keys and strict-acceptance rationale; the [V2-to-V3 specification](../../../../packages/session/session-format-v2-to-v3/README.md#canonical-envelopes) owns historical conversion. This note retains ordered-projection ownership and replacement rationale.
The fields are serialized as top-level JSON properties. JSONL preserves placement and source-event references without a separate column mapping. The [V3 canonical-envelope decision](2026-09-06-v3-canonical-session-envelopes.md) owns exact replacement keys and strict-acceptance rationale; the [V2-to-V3 specification](../../../../packages/session/session-format-v2-to-v3/README.md#canonical-envelopes) owns historical conversion. This note retains ordered-projection ownership and replacement rationale.
### Crash recovery
@@ -37,7 +37,7 @@ surface 元数据仅属于四种 surface 事件类型(`system/message`、`user
### 持久化
这些字段作为顶层 JSON 属性序列化。JSONL 无需单独列映射即可保留位置与来源。[V3 规范信封决策](2026-09-06-v3-canonical-session-envelopes.zh.md)负责精确替换键与严格准入依据;[V2 到 V3 规范](../../../../packages/session/session-format-v2-to-v3/README.zh.md#canonical-envelopes)负责历史转换。本文继续负责有序投影的所有权与替换依据。
这些字段作为顶层 JSON 属性序列化。JSONL 无需单独列映射即可保留位置与源事件引用。[V3 规范信封决策](2026-09-06-v3-canonical-session-envelopes.zh.md)负责精确替换键与严格准入依据;[V2 到 V3 规范](../../../../packages/session/session-format-v2-to-v3/README.zh.md#canonical-envelopes)负责历史转换。本文继续负责有序投影的所有权与替换依据。
### 崩溃恢复
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-05-prompt-variables-and-tool-guidance-ownership.md
2026-07-05-prompt-variables-and-tool-guidance-ownership.md: 35bb7c6fabc85ae6f93bdbb67e13910eea627ca3
2026-07-05-prompt-variables-and-tool-guidance-ownership.zh.md: 53d69cf45c02f6534334561b626d2c2ae6087c05
2026-07-05-prompt-variables-and-tool-guidance-ownership.md: 8f5542cb09930e62fd1e26960166fbdd6d3f6745
2026-07-05-prompt-variables-and-tool-guidance-ownership.zh.md: 348ae36cf6549d412adab1d3cdb5cc03f0badd01
@@ -26,7 +26,7 @@ The assembled system prompt had four defects, all of one family: facts the harne
### Prompt variables
Plugins register `{{name}}` values through `ctx.systemPrompt.variable(name, provider)`. Assembly resolves them into the waterfall-visible variable map. Rendering rejects unknown own-property references, registered providers that return `undefined`, malformed complete references, and unbalanced references that still contain a closing `}}`; a lone unmatched `{{` remains prose, and substituted values are not rescanned. Registration rejects invalid or duplicate variable names, and section names are unique.
Plugins register `{{name}}` values through `ctx.systemPrompt.variable(name, provider)`. Assembly resolves them into the waterfall-visible variable map. Rendering rejects unknown own-property references, registered providers that return `undefined`, malformed complete references, and unbalanced references that still contain a closing `}}`; a lone unmatched `{{` remains prose, and substituted values are not rescanned. Registration rejects invalid or duplicate variable names, and section names are unique. Sections may set `interpolate: false` to preserve generated documents literally; `tools:sdk` does so because tool descriptions and schemas may document their own `{{…}}` syntax.
`dsh-agent-loop` registers the two built-ins, both pure projections of the context agent: `model` (= `options.model`) and `cwd` (= `session.header.cwd`). The example personas write `powered by the {{model}} model` — the model name is stated once, in the `model:` config key. `{{cwd}}` is demonstrated in the ACP example only: every ACP session carries the client's cwd, while config-pre-created stdio agents have none (a persona claiming `{{cwd}}` there fails the turn — by design). The variables stay on the loop plugin (unlike the sections below): they are runtime facts of the agents THIS loop drives, and a replacement loop supplies its own.
@@ -47,7 +47,7 @@ Per-tool semantics and selection guidance live in tool descriptions. Prompt sect
- **The loop composes an identity line itself** — hardcodes model-facing prose in the one package that must stay thin ("plugins, not loop changes"), and outside the section pipeline it would be a second composition path. (The identity DOES ship as a code literal — but as an ordinary section registered by `dsh-system-prompt`, whose `system-prompt/assemble` waterfall remains the escape valve for a deployment that must drop it.)
- **Inject the model name via the `agent/request` waterfall** — prompt text would be composed in two places and the earlier rendered persona could disagree with the final routed header. The request plugin that owns late routing must also own any earlier prompt claim about that model.
- **Hand-write the model name in each persona** — duplicates the `model:` key one line above and silently lies after a config edit; the exact disease this decision cures.
- **Lenient interpolation (leave unknown refs verbatim, or substitute empty)** — a typo ships `{{modle}}` (or a hole) to the model and nobody notices until transcript review.
- **Lenient interpolation (leave unknown refs verbatim, or substitute empty)** — a typo ships `{{modle}}` (or a hole) to the model and nobody notices until transcript review. Leaving only unknown names unchanged would still substitute registered names inside tool documentation.
- **Per-instance subagent wording in config** — returns model-facing prose to every deployment × instance, reviving the hand-written-guidance-in-leaf-YAML drift. **Keying wording off the provider NAME** — `providerName` is itself config, so a renamed provider silently gets the wrong words.
- **Resolving the provider at `apply` time (a load-order requirement)** and **section-only subagent wording (lazily resolved at assemble)** — the alternatives to the provider-lifecycle events; both rejected in [the provider-lifecycle-events Agent Note](../../archived/architecture/2026-07-05-subagent-provider-lifecycle-events.md).
@@ -60,7 +60,7 @@ Per-tool semantics and selection guidance live in tool descriptions. Prompt sect
- The tui-agent prompt renders identity, persona with the interpolated model, then fs/shell/web guidance through one assembly path.
- Fork and fresh subagent descriptions reflect whether the provider inherits completed conversation turns; the tool appears, disappears, and is reworded with provider lifecycle changes.
- Unknown, valueless, malformed, or unbalanced variable references name the section and throw; duplicate section, variable, and tool registrations also throw.
- In interpolated sections, unknown, valueless, malformed, or unbalanced variable references name the section and throw; duplicate section, variable, and tool registrations also throw.
- Snapshot replay is prompt-independent: it keys recorded chunk streams by turn and step without comparing the outgoing request.
## Consequences
@@ -69,4 +69,4 @@ Per-tool semantics and selection guidance live in tool descriptions. Prompt sect
- `{{model}}` reflects `AgentOptions.model` at assembly time. A plugin that switches models in the `agent/request` waterfall makes the prompt's claim stale for that step, and one that SUPPLIES the model there (options.model unset — the loop's documented fallback) leaves the variable valueless at render, failing a `{{model}}` persona before the waterfall runs. Both have the same remedy, and it is the ownership rule itself: the plugin that owns the late-bound model fact states it early on the `system-prompt/assemble` waterfall (`assembly.variables['model'] = …`) — one owner, both statements; a loop test pins the supply path end-to-end. Accepted.
- While a bound provider is absent (not yet activated, unloaded, mid-HMR-reload), the subagent tool does not exist and a model request in that window simply lacks it. That is the honest state — the alternative was a registered tool whose description or execution could not be trusted.
- Strictness means a persona can fail a turn at render (e.g. `{{cwd}}` on a cwd-less session). The failure is contained — the turn ends `error`, the loop survives — and it is an authoring error we WANT loud.
- No escape syntax for a literal `{{name}}` in prompt prose yet; add one if a real prompt ever needs it.
- Inline escapes remain unsupported in interpolated text; literal sections need no escaping. PTC unit tests cover both modes and runtime languages, and the recorded `ptc-turn` scenario preserves tool-template examples in the model-visible prompt.
@@ -26,7 +26,7 @@ Status: implemented
### 提示词变量
插件通过 `ctx.systemPrompt.variable(name, provider)` 注册 `{{name}}` 值。组装过程将它们解析到 waterfall 可见的变量映射中。渲染阶段拒绝以下情况:引用未知的自有属性、已注册的提供方返回 `undefined`、格式错误的完整引用、以及仍包含闭合 `}}` 的不平衡引用;孤立的未匹配 `{{` 保留为行文,替换后的值不会被重新扫描。注册阶段拒绝无效或重复的变量名,section 名称也必须唯一。
插件通过 `ctx.systemPrompt.variable(name, provider)` 注册 `{{name}}` 值。组装过程将它们解析到 waterfall 可见的变量映射中。渲染阶段拒绝以下情况:引用未知的自有属性、已注册的提供方返回 `undefined`、格式错误的完整引用、以及仍包含闭合 `}}` 的不平衡引用;孤立的未匹配 `{{` 保留为行文,替换后的值不会被重新扫描。注册阶段拒绝无效或重复的变量名,section 名称也必须唯一。段可设置 `interpolate: false` 来原样保留生成的文档;`tools:sdk` 使用此设置,因为工具描述和 schema 可能会介绍自身的 `{{…}}` 语法。
`dsh-agent-loop` 注册两个内置变量,均为上下文 agent 的纯投影:`model`(= `options.model`)和 `cwd`(= `session.header.cwd`)。示例 persona 写 `powered by the {{model}} model`——模型名称只在 `model:` 配置键中声明一次。`{{cwd}}` 仅在 ACP 示例中演示:每个 ACP 会话携带客户端的 cwd,而配置预创建的 stdio agent 没有 cwd(在那里声称 `{{cwd}}` 的 persona 会导致该轮次失败——这是有意为之)。变量留在 loop 插件上(不同于下面的 section):它们是本循环驱动的 agent 的运行时事实,替换循环自行提供自己的变量。
@@ -47,7 +47,7 @@ Status: implemented
- **循环自行组合一行 identity 文本**:在必须保持精简的那个包(「用插件,不改循环」)中硬编码面向模型的行文,且在 section 流水线之外构成第二条组合路径。(identity 确实以代码字面量交付——但作为 `dsh-system-prompt` 注册的普通 section,其 `system-prompt/assemble` waterfall 仍是部署需要移除它时的逃生阀。)
- **通过 `agent/request` waterfall 注入模型名称**:提示词文本会在两处组合,更早渲染的 persona 也可能与最终已路由 header 不一致。拥有延迟路由的请求插件还必须拥有该模型在提示词中更早出现的声明。
- **在每个 persona 中手写模型名称**:与上方一行的 `model:` 键重复,配置修改后静默失实;正是本决策要治愈的病症。
- **宽松插值(未知引用保留原样或替换为空)**:一个拼写错误 `{{modle}}`(或一个空洞)会被发送给模型,直到 transcript(文本记录)审查时才会被发现。
- **宽松插值(未知引用保留原样或替换为空)**:一个拼写错误 `{{modle}}`(或一个空洞)会被发送给模型,直到 transcript(文本记录)审查时才会被发现。仅保留未知名称仍会替换工具文档中的已注册名称。
- **在配置中为每个 subagent 实例编写措辞**:面向模型的行文回到每个部署 × 实例中,重蹈在 leaf YAML 中手写指导的漂移。**根据提供方名称选择措辞**:`providerName` 本身是配置,重命名提供方后会静默获得错误的措辞。
- **在 `apply` 时解析提供方(加载顺序要求)**与**仅用 section 承载 subagent 措辞(在 assemble 时惰性解析)**:提供方生命周期事件的替代方案;两者均在[提供方生命周期事件 Agent Note](../../archived/architecture/2026-07-05-subagent-provider-lifecycle-events.md)中被否决。
@@ -60,7 +60,7 @@ Status: implemented
- tui-agent 的提示词通过一条组装路径依次渲染 identity、带插值模型名的 persona,然后是 fs/shell/web 指导。
- fork 和 fresh subagent 的描述反映提供方是否继承已完成的对话轮次;工具随提供方生命周期变化而出现、消失和重新措辞。
- 未知、无值、格式错误或不平衡的变量引用会指明 section 名称并抛出异常;重复的 section、变量和工具注册同样抛出异常。
- 在启用插值的段中,未知、无值、格式错误或不平衡的变量引用会指明 section 名称并抛出异常;重复的 section、变量和工具注册同样抛出异常。
- 快照回放与提示词无关:它按轮次和步骤索引已记录的分片流,不比较发出的请求。
## 后果
@@ -69,4 +69,4 @@ Status: implemented
- `{{model}}` 在组装时反映 `AgentOptions.model`。如果一个插件在 `agent/request` waterfall 中切换模型,提示词对该步骤的声明就会过时;如果一个插件在那里提供模型(options.model 未设置——循环文档中记载的回退路径),变量在渲染时无值,包含 `{{model}}` 的 persona 会在 waterfall 运行前失败。两者的补救方式相同,就是归属规则本身:拥有延迟绑定模型事实的插件在 `system-prompt/assemble` waterfall 上提前声明它(`assembly.variables['model'] = …`)——一个归属方,两处声明;一个循环测试端到端固定了 supply 路径。已接受。
- 当一个已绑定的提供方不存在时(尚未激活、已卸载、HMR(热模块替换)重载中),subagent 工具不存在,该窗口内的模型请求中不会包含它。这是诚实的状态——替代方案是注册一个 description 或执行都不可信的工具。
- 严格性意味着 persona 可能在渲染时导致轮次失败(例如在无 cwd 的会话上使用 `{{cwd}}`)。失败是受控的——该轮次以 `error` 结束,循环存活——且这是一个我们希望明确暴露的撰写错误。
- 目前没有在提示词行文中转义字面 `{{name}}` 的语法;如果真实提示词确实需要,再行添加。
- 插值文本仍不支持行内转义;字面文本段无需转义。PTC 单元测试覆盖两种模式和运行时语言,录制的 `ptc-turn` 场景在模型可见的提示词中保留工具模板示例。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-05-reconstructable-requests.md
2026-07-05-reconstructable-requests.md: 2f88675a75a72e7fbf105dfbf4f337a4dd80948a
2026-07-05-reconstructable-requests.zh.md: 90008502a6651e38c142b7fb88052c05d46dea76
2026-07-05-reconstructable-requests.md: f47d0938b91edf6c9b544223ac44925f02dd8c1f
2026-07-05-reconstructable-requests.zh.md: d2499a6264a3e6bebf31944d8eb056265e082c68
@@ -22,9 +22,9 @@ Prefix-cache stability is corollary #1, not the headline: an append-only log pro
**Messages.** `Session.deriveMessages()` is cached: each surface entry is projected exactly once, when first seen, through the public per-event function `deriveEventMessage(event)`; a surface rewrite (a compaction `replace` — `SurfaceManager.replaceGeneration`) rebuilds. Callers get a fresh array per call over shared, deep-frozen messages: mutating logged history through a projection is unrepresentable (it throws), replacing the old clone-per-call isolation. External reconstructors fold the same public function over a log prefix, so no two paths can disagree.
`EpochHeader` records the request's non-history state: call config and tool schemas. Writers omit `tools: []` and `adapterDefaults: {}`; current acceptance rejects those fields and any `header.system`, rather than repairing them. Whitespace-only system-message content, `config.stop: []`, and nested extensions remain intact. The [V3 canonical-envelope decision](2026-09-06-v3-canonical-session-envelopes.md) owns historical conversion. The rendered system prompt is derived history — the `system/message` event at surface node 0, per the [surface-node Agent Note](2026-09-02-system-prompt-as-surface-node.md) — so a prompt change is a surface replacement rather than a header change. Adapter-supplied effort and token defaults retain their `adapterDefaults` provenance; a Web model selection restored from the log omits an adapter-owned effort so the next resolution cannot reclassify the same effective config as an explicit selection and a false change. `request/header` always writes a full snapshot: the first loop instance uses reason `initial`, later instances use `resume`, an in-instance change uses `change`, and an unchanged envelope beginning an explicitly declared message series or following a surface replacement uses `series`. A `change` snapshot carries `startsSeries: true` when the changed request also starts a series, preserving the two independent facts without a duplicate header. Ordinary append-only later Turns, further same-series Steps, and retries inherit the latest snapshot. `foldRequestHeader` selects the latest snapshot. Legacy `request/header-delta` events and the removed `fallback` reason are rejected when appended or loaded.
`EpochHeader` records the request's non-history state: call config and tool schemas. Writers omit `tools: []` and `adapterDefaults: {}`; current acceptance rejects those fields and any `header.system`, rather than repairing them. Whitespace-only system-message content, `config.stop: []`, and nested extensions remain intact. The [V3 canonical-envelope decision](2026-09-06-v3-canonical-session-envelopes.md) owns historical conversion. The rendered system prompt is derived history — the `system/message` event at surface node 0, per the [surface-node Agent Note](2026-09-02-system-prompt-as-surface-node.md) — so a prompt change is a surface replacement rather than a header change. Adapter-supplied effort and token defaults retain their `adapterDefaults` ownership; a Web model selection restored from the log omits an adapter-owned effort so the next resolution cannot reclassify the same effective config as an explicit selection and a false change. `request/header` always writes a full snapshot: the first loop instance uses reason `initial`, later instances use `resume`, an in-instance change uses `change`, and an unchanged envelope beginning an explicitly declared message series or following a surface replacement uses `series`. A `change` snapshot carries `startsSeries: true` when the changed request also starts a series, preserving the two independent facts without a duplicate header. Ordinary append-only later Turns, further same-series Steps, and retries inherit the latest snapshot. `foldRequestHeader` selects the latest snapshot. Legacy `request/header-delta` events and the removed `fallback` reason are rejected when appended or loaded.
Each proposed step first claims its inbox batch, assembles the system prompt and tools, projects the rendered prompt against the surviving `system/message` node, and runs `agent/pre-step`. Rejection opens no step; enter opens `step/start`, commits a changed prompt as the `system/message` append or node-0 replacement, records the final message batch as `user/message` events, and may use `startsRequestSeries: true` to declare a distinct series. `agent/request` may replace only the frozen call-config seed. The loop records the owed initial, resume, change, or series full snapshot, builds `GenerateOptions` from derived messages (system message first) and that header with no `system` field, and freezes it while leaving `AbortSignal` live. The [request-freeze provenance decision](../simplification/2026-09-06-agent-request-freeze-provenance.md) owns reuse of completed message freezes and per-request local header freezing. The first call config starts from explicit `AgentOptions`, preserving fork overrides and resume reconfiguration; later calls start from the folded header.
Each proposed step first claims its inbox batch, assembles the system prompt and tools, projects the rendered prompt against the surviving `system/message` node, and runs `agent/pre-step`. Rejection opens no step; enter opens `step/start`, commits a changed prompt as the `system/message` append or node-0 replacement, records the final message batch as `user/message` events, and may use `startsRequestSeries: true` to declare a distinct series. `agent/request` may replace only the frozen call-config seed. The loop records the owed initial, resume, change, or series full snapshot, builds `GenerateOptions` from derived messages (system message first) and that header with no `system` field, and freezes it while leaving `AbortSignal` live. The [request-freeze evidence decision](../simplification/2026-09-06-agent-request-freeze-evidence.md) owns reuse of completed message freezes and per-request local header freezing. The first call config starts from explicit `AgentOptions`, preserving fork overrides and resume reconfiguration; later calls start from the folded header.
**The open step is the reconstruction boundary.** Its entered `user/message` batch and any newly written `request/header` precede request dispatch. Injection after the atomic claim joins a later request, while a listener that must affect this request returns messages through `agent/pre-step`. Header reconstruction selects the step's `request/header`, or carries the prior snapshot when no new header is written.
@@ -24,7 +24,7 @@ Status: implemented
`EpochHeader` 记录请求的非历史状态:调用配置和工具 schema。写入方省略 `tools: []` 与 `adapterDefaults: {}`;当前接纳拒绝这些字段以及任何 `header.system`,而不修复它们。仅含空白的系统消息内容、`config.stop: []` 与嵌套扩展保持原样。[V3 规范信封决策](2026-09-06-v3-canonical-session-envelopes.zh.md)负责历史转换。渲染后的系统提示词是派生历史——surface 第 0 号节点上的 `system/message` 事件,见[surface 节点 Agent Note](2026-09-02-system-prompt-as-surface-node.zh.md)——因此提示词变更是 surface 替换而不是 header 变更。适配器提供的推理强度与 token 默认值会保留其 `adapterDefaults` 来源信息;Web 从日志恢复模型选择时会省略适配器持有的推理强度,因此下一次解析不会把相同的有效配置重新归类为显式选择并产生虚假变更。`request/header` 始终写入完整快照:首个循环实例使用 reason `initial`,后续实例使用 `resume`,实例内变更使用 `change`,内容未变的封装显式开启消息序列或跟随表层替换时使用 `series`。如果发生变化的请求同时开启序列,`change` 快照会携带 `startsSeries: true`,无需重复 header 即可保留这两个独立事实。普通的仅追加后续 Turn、同一序列内后续的 Step 与重试沿用最新快照。`foldRequestHeader` 选择最新快照。旧的 `request/header-delta` 事件和已移除的 `fallback` reason 在追加或加载时都会被拒绝。
每个拟议步骤先领取其 inbox 批次,组装系统提示词与工具,把渲染后的提示词与存活的 `system/message` 节点比对投影,再运行 `agent/pre-step`。reject 不打开步骤;enter 打开 `step/start`,把变化的提示词作为 `system/message` 追加或第 0 号节点替换提交,把最终消息批次记录为 `user/message` 事件,并可使用 `startsRequestSeries: true` 声明独立序列。`agent/request` 只能替换冻结的调用配置种子。循环记录所需的 initial、resume、change 或 series 完整快照,从派生消息(系统消息在先)与该不含 `system` 字段的 header 构建 `GenerateOptions`,冻结请求但保持 `AbortSignal` 活跃。[请求冻结来源证明决策](../simplification/2026-09-06-agent-request-freeze-provenance.zh.md)拥有消息完整冻结的复用规则和每次请求的本地 header 冻结规则。首次调用配置从显式的 `AgentOptions` 出发,保留 fork 覆盖和恢复重配置;后续调用从折叠后的 header 出发。
每个拟议步骤先领取其 inbox 批次,组装系统提示词与工具,把渲染后的提示词与存活的 `system/message` 节点比对投影,再运行 `agent/pre-step`。reject 不打开步骤;enter 打开 `step/start`,把变化的提示词作为 `system/message` 追加或第 0 号节点替换提交,把最终消息批次记录为 `user/message` 事件,并可使用 `startsRequestSeries: true` 声明独立序列。`agent/request` 只能替换冻结的调用配置种子。循环记录所需的 initial、resume、change 或 series 完整快照,从派生消息(系统消息在先)与该不含 `system` 字段的 header 构建 `GenerateOptions`,冻结请求但保持 `AbortSignal` 活跃。[请求冻结证据决策](../simplification/2026-09-06-agent-request-freeze-evidence.zh.md)拥有消息完整冻结的复用规则和每次请求的本地 header 冻结规则。首次调用配置从显式的 `AgentOptions` 出发,保留 fork 覆盖和恢复重配置;后续调用从折叠后的 header 出发。
**已打开步骤是重建边界。** 进入步骤的 `user/message` 批次与任何新写入的 `request/header` 都位于请求分派之前。原子领取后发生的注入加入后续请求;必须影响本次请求的监听器则通过 `agent/pre-step` 返回消息。header 重建选择该步骤的 `request/header`,或在无新 header 写入时沿用前一个快照。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-08-tool-output-spill-files.md
2026-07-08-tool-output-spill-files.md: a80b6cdcb0e731a687d511d38ea288a68a298173
2026-07-08-tool-output-spill-files.md: fa6697aad63c97b0ef580fe1ca46e50541e3d682
2026-07-08-tool-output-spill-files.zh.md: a995eafc7878b342a2164c5116da1f43ada791a5
@@ -22,7 +22,7 @@ A thin spill storage seam plus a default spill policy plugin, in a new `packages
| `@deepseek-ai/dsh-spill-local` | Local backend: private, session-scoped file storage on the host filesystem. |
| `@deepseek-ai/dsh-spill-policy` | Tool-result policy plugin: wraps final text results after dispatch and replaces oversized results with a retained preview plus a spill locator. |
The tool-result Consumer is `dsh-spill-policy`, which consumes final tool results through the `tools/post-execute` waterfall. The model follows the backend-supplied retrieval hint for the returned locator. [Session-reference spill reuse](../bug-fix/2026-09-05-session-reference-spill-reuse.md) adds a direct storage consumer with separate preview, provenance, and failure semantics; it does not change the tool-result policy.
The tool-result Consumer is `dsh-spill-policy`, which consumes final tool results through the `tools/post-execute` waterfall. The model follows the backend-supplied retrieval hint for the returned locator. [Session-reference spill reuse](../bug-fix/2026-09-05-session-reference-spill-reuse.md) adds a direct storage consumer with separate preview, source-description, and failure semantics; it does not change the tool-result policy.
### Spill seam
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-10-single-file-executable-sdk-runtime-distribution.md
2026-07-10-single-file-executable-sdk-runtime-distribution.md: 5533c38d635d04c8799658fc1f6bcec8543d1dab
2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md: 05800082034f0d1bd5034fc5bb17f09356d84dd9
2026-07-10-single-file-executable-sdk-runtime-distribution.md: 665364e6d39a78f7ac497198f18ed9aa160a4cbd
2026-07-10-single-file-executable-sdk-runtime-distribution.zh.md: 8b39df71c615dc59f3ef1bf622a5736a70b085b3
@@ -40,13 +40,13 @@ Inside the exe's VFS sits a **real package tree in build-artifact form** (each p
The deploy root is [`python/sdk-runtime/package.json`](../../../../python/sdk-runtime/package.json) (`dsh-python-runtime-closure`, a pnpm workspace member and a zero-code pure dependency manifest) — the unified source of truth for "which plugins the exe ships" and "what the Python runtime distributes". Adding a plugin to the exe = adding one dependency line to the manifest and repackaging. [`scripts/verify-runtime-closure.ts`](../../../../scripts/verify-runtime-closure.ts) reads every shipped `packages/preset/agent-presets/presets/*/agent.cordis.yml`, evaluates `disabled` conditions that compare `process.platform` for every target in `python/sdk-runtime/platforms.json`, and requires each active workspace plugin at the runtime root through an explicit `workspace:` dependency. It also traverses every workspace package covered by that manifest and requires every non-optional workspace peer, reporting the complete preset or referencing-package → missing-dependency chain; unknown platform conditions remain active so a plugin cannot be omitted by an unsupported expression. `pnpm run hygiene`, CI static, and the single-exe build run it before packaging. Deploy also packs by each package's `files`, so the shared chunks tsdown splits out must be covered by `files`.
The deploy root includes `@deepseek-ai/dsh-mcp-client` as an explicitly supported custom-configuration plugin even though no shipped preset mounts it. An external config can therefore connect to user-supplied stdio and Streamable HTTP MCP servers and register their tools; the distribution does not carry those servers or extend the bridge to MCP Resources and Prompts. The executable and installed-wheel smokes start a temporary stdio server, discover its tool, and complete one model-requested call.
The deploy root includes `@deepseek-ai/dsh-mcp-client` as an explicitly supported custom-configuration plugin even though no shipped preset mounts it. An external config can therefore connect to user-supplied stdio and Streamable HTTP MCP servers and register their tools; the distribution does not carry those servers. The optional resource service provides MCP Resources; MCP Prompts remain unsupported. The executable and installed-wheel smokes start a temporary stdio server, discover its tool, and complete one model-requested call.
### Build pipeline and artifacts
[`scripts/build-exe-for-python-sdk.ts`](../../../../scripts/build-exe-for-python-sdk.ts): runtime closure verification → `pnpm run build` → (after clearing) `pnpm --filter dsh-python-runtime-closure deploy --legacy --prod --config.node-linker=hoisted --config.auto-install-peers=false --config.link-workspace-packages=true` **directly into** `python/sdk-runtime/src/deepseek_harness_runtime/runtime/node/`, including [`python/sdk-runtime/runtime-bootstrap.mjs`](../../../../python/sdk-runtime/runtime-bootstrap.mjs) as the carrier-root `runtime-bootstrap.mjs` → restore direct workspace packages omitted by legacy deploy and reject any remaining manifest gap → replace staged dependency symlinks with their target bytes, remove package-manager `.bin` links, and fail if any symlink remains → verify the deployed bootstrap and inject pkg configuration with that bin plus assets covering dynamic profile, bundle, frontend, preset, native-library, and configuration reads → stage the target `node-pty` addon → invoke `pkg --sea` once per target → write `deepseek-harness-sdk-runtime-<platform>-<arch>` under `dist-exe/` and copy it into the runtime directory. The Python runtime owns that bootstrap. It calls the public CLI export for ordinary launches and dispatches a provider-private selection to the same `@deepseek-ai/dsh-subprocess-local/runner` core without changing CLI grammar or adding another executable; the [native-containment decision](2026-08-28-subprocess-native-containment.md) owns that private path. Linux CI rebuilds `pty.node` inside the matching manylinux 2.28 container because legacy deploy omits that install side effect. Every target copies its native `@vscode/ripgrep` binary beside the executable as the required `-rg` sidecar; pkg runtimes select that sidecar through `process.pkg`, while ordinary Node execution uses `@vscode/ripgrep` directly. macOS uses its target prebuild and also emits the required `-spawn-helper`. All four deploy flags are grounded in measurement: `--legacy` is the mandatory path with inject-workspace-packages off; hoisted gives pkg a stable single-instance layout that the explicit materialization pass makes symlink-free; disabling automatic peer installation prevents undeclared peers from expanding the closure; link-workspace-packages selects direct workspace dependencies. [`pnpm-workspace.yaml`](../../../../pnpm-workspace.yaml) overrides the transitive `@deepseek-ai/cosmokit` and `@deepseek-ai/schemastery` semver requests to the pinned vendor sources so legacy deploy never resolves those unpublished names from a registry.
[`scripts/build-exe-for-python-sdk.ts`](../../../../scripts/build-exe-for-python-sdk.ts): runtime closure verification → `pnpm run build` → (after clearing) `pnpm --filter dsh-python-runtime-closure deploy --legacy --prod --config.allow-unused-patches=true --config.node-linker=hoisted --config.auto-install-peers=false --config.link-workspace-packages=true` **directly into** `python/sdk-runtime/src/deepseek_harness_runtime/runtime/node/`, including [`python/sdk-runtime/runtime-bootstrap.mjs`](../../../../python/sdk-runtime/runtime-bootstrap.mjs) as the carrier-root `runtime-bootstrap.mjs` → restore direct workspace packages omitted by legacy deploy and reject any remaining manifest gap → replace staged dependency symlinks with their target bytes, remove package-manager `.bin` links, and fail if any symlink remains → verify the deployed bootstrap and inject pkg configuration with that bin plus assets covering dynamic profile, bundle, frontend, preset, native-library, and configuration reads → stage the target `node-pty` addon → invoke `pkg --sea` once per target → write `deepseek-harness-sdk-runtime-<platform>-<arch>` under `dist-exe/` and copy it into the runtime directory. The Python runtime owns that bootstrap. It calls the public CLI export for ordinary launches and dispatches a provider-private selection to the same `@deepseek-ai/dsh-subprocess-local/runner` core without changing CLI grammar or adding another executable; the [native-containment decision](2026-08-28-subprocess-native-containment.md) owns that private path. Linux CI rebuilds `pty.node` inside the matching manylinux 2.28 container because legacy deploy omits that install side effect. Every target copies its native `@vscode/ripgrep` binary beside the executable as the required `-rg` sidecar; pkg runtimes select that sidecar through `process.pkg`, while ordinary Node execution uses `@vscode/ripgrep` directly. macOS uses its target prebuild and also emits the required `-spawn-helper`. The deploy flags preserve the measured runtime layout: `--legacy` is the mandatory path with inject-workspace-packages off; hoisted gives pkg a stable single-instance layout that the explicit materialization pass makes symlink-free; disabling automatic peer installation prevents undeclared peers from expanding the closure; link-workspace-packages selects direct workspace dependencies. [`pnpm-workspace.yaml`](../../../../pnpm-workspace.yaml) overrides the transitive `@deepseek-ai/cosmokit` and `@deepseek-ai/schemastery` semver requests to the pinned vendor sources so legacy deploy never resolves those unpublished names from a registry. Production deployment permits unused workspace patches because development-only tools can be absent from its closure; included-package patch failures still fail deployment. Workspace installs retain strict unused-patch validation.
CI: [`.github/workflows/build-exe-for-python-sdk.yml`](../../../../.github/workflows/build-exe-for-python-sdk.yml) runs [installed-wheel validation](../testing/2026-08-23-installed-python-wheel-black-box-ci.md) on Linux/Windows x64 for pull requests and Linux ARM64 plus both macOS architectures for master pushes. The [public publication workflow](../../archived/process/2026-08-11-python-publication-workflow.md) calls it for all five targets; `workflow_dispatch` can still select a subset. Native builds run on linux-x64 / linux-arm64 (`ubuntu-24.04-arm`) / macos-arm64 / macos-x64 (`macos-15-intel`) / win-x64 (`windows-2025`), with `~/.pkg-cache` cached where applicable, and pkg handles macOS ad-hoc signing. Each leg installs the release-shaped SDK and runtime wheels into a clean venv outside the checkout, proves their package and executable provenance, then drives the complete keyless scenario set through the public SDK and direct NDJSON JSON-RPC. Trusted pull requests and master pushes additionally run a real DeepSeek two-turn tool smoke on their selected targets; fork and Dependabot heads receive no key. Linux inspects the executable and native addon's GLIBC requirements and runs an additional manylinux 2.28 smoke, while macOS checks the runtime, ripgrep, and PTY helper architectures and verifies that all three deployment targets fit the wheel tag. A full five-target run retains six artifacts, each containing one release file: the platform-independent SDK wheel and five native runtime wheels; a subset dispatch retains the SDK wheel and selected runtime wheels. Bare executables and source bundles remain intermediate test inputs. [`.gitlab-ci.yml`](../../../../.gitlab-ci.yml) accepts `python-v<repository-version>` tag pipelines whose version matches the root `package.json`, builds one SDK wheel and five native runtime wheels, then a single serialized job checks and publishes all six to the project PyPI registry. The [`python/sdk-runtime` README](../../../../python/sdk-runtime/README.md) owns the Windows target and the explicit exclusion of Windows arm64.
CI: [`.github/workflows/build-exe-for-python-sdk.yml`](../../../../.github/workflows/build-exe-for-python-sdk.yml) runs [installed-wheel validation](../testing/2026-08-23-installed-python-wheel-black-box-ci.md) on Linux/Windows x64 for pull requests and Linux ARM64 plus both macOS architectures for master pushes. The [public publication workflow](../../archived/process/2026-08-11-python-publication-workflow.md) calls it for all five targets; `workflow_dispatch` can still select a subset. Native builds run on linux-x64 / linux-arm64 (`ubuntu-24.04-arm`) / macos-arm64 / macos-x64 (`macos-15-intel`) / win-x64 (`windows-2025`), with `~/.pkg-cache` cached where applicable, and pkg handles macOS ad-hoc signing. Each leg installs the release-shaped SDK and runtime wheels into a clean venv outside the checkout, verifies their package and executable origin, then drives the complete keyless scenario set through the public SDK and direct NDJSON JSON-RPC. Trusted pull requests and master pushes additionally run a real DeepSeek two-turn tool smoke on their selected targets; fork and Dependabot heads receive no key. Linux inspects the executable and native addon's GLIBC requirements and runs an additional manylinux 2.28 smoke, while macOS checks the runtime, ripgrep, and PTY helper architectures and verifies that all three deployment targets fit the wheel tag. A full five-target run retains six artifacts, each containing one release file: the platform-independent SDK wheel and five native runtime wheels; a subset dispatch retains the SDK wheel and selected runtime wheels. Bare executables and source bundles remain intermediate test inputs. [`.gitlab-ci.yml`](../../../../.gitlab-ci.yml) accepts `python-v<repository-version>` tag pipelines whose version matches the root `package.json`, builds one SDK wheel and five native runtime wheels, then a single serialized job checks and publishes all six to the project PyPI registry. The [`python/sdk-runtime` README](../../../../python/sdk-runtime/README.md) owns the Windows target and the explicit exclusion of Windows arm64.
### Python SDK distribution: two carriers, exe for production, node for development
@@ -60,13 +60,13 @@ The Python client launches the packaged `dsh` command with the selected profile
`dsh-python-runtime-closure` is the private deploy manifest and `deepseek-harness-sdk-runtime-<platform>-<arch>` is the executable family. The wire `serverInfo.name` is `deepseek-harness-sdk-runtime`; the Python distribution names are `deepseek-harness-sdk` / `deepseek-harness-runtime-bin`, while the import modules are `deepseek_harness` / `deepseek_harness_runtime`.
## Disposition of worker-style plugins
## Packaged workflow and code execution
`dsh-workflow-worker-thread` and `dsh-code-runtime-worker-thread` are supported inside the exe. Their built hosts convert the sibling `lib/worker.cjs` URL with `fileURLToPath()` and pass the resulting filesystem string to `Worker`, which is the form pkg's Worker hook resolves inside the VFS. The worker entries are CommonJS because that hook compiles VFS worker files as CommonJS. The workflow engine keeps its data-URL bootstrap for unbuilt source execution; only its built sibling entry uses the filesystem string. The custom-config executable smoke loads both backends, invokes a real `run_code` call and a zero-agent `workflow` call, and requires each worker to return `42` from inside pkg's VFS.
`dsh-workflow-ptc` sends its self-contained guest program through `dsh-ptc-runtime-node`. The shared provider starts a managed child through the executable's private Node bootstrap dispatch, with a dedicated control channel; no separate workflow worker asset is required. The [sandboxed Node decision](2026-09-11-sandboxed-node-ptc-runtime.md) owns process and policy lifetime, and [workflow sandbox reuse](2026-09-13-workflow-ptc-sandbox-reuse.md) owns the workflow adapter.
## Testing
The verification surface has three tiers. Mechanism tier: the measured conclusions for the `--sea` chain are embedded in the Decision sections (ESM dynamic import inside the VFS, single cordis instance, fail-loud config chain, `node:sqlite`, macOS ad-hoc signing runs). SDK tier: the complete keyless pytest suite covers the client protocol against a fake runtime peer, subprocess cleanup, absolute cwd propagation, dual-carrier launch, and carrier resolution; root CI runs it on Python 3.10. End-to-end tier: every platform build installs both wheels into a clean venv outside the checkout, proves matching versions and installed module/executable locations, then completes turns against a mock endpoint through the default SDK path, a custom config, the checked-in standalone minimal composition, and the direct binary protocol, with final text and JSONL checked. The minimal run asserts its exact system prompt and two-tool catalog, retains Bash state across calls, and invokes the editor. The custom config additionally drives `run_code` and a zero-agent `workflow` through their real worker files inside the packaged VFS. The filesystem-search scenario requires the model to call both `glob` and `grep` through the target-native `-rg` sidecar. The spawn-node scenario drives the platform shell tool through a command starting with `node` and requires the machine's own Node version in the tool result with no `PKG_EXECPATH` in the child environment, pinning the packaged runtime against a pkg upgrade that re-records the child-process patch. The MCP scenario starts a temporary external stdio server, deliberately delays its initial `tools/list` response, then immediately starts the first SDK prompt; the prompt must see and call the discovered tool, proving that `initialize` is a real Loader-settlement readiness boundary rather than a timing sleep. The same installed run compares a committed executable-specific snapshot through the Python SDK: a keyless scripted model mounts a Cordis plugin that registers a tool, invokes that tool from `run_code`, runs a direct spawn subagent and a workflow that starts a second spawn child, then unmounts the plugin. The fixture explicitly disables its unused bundled Bash and local skill discovery so its tool set does not depend on repository-external state, and the comparison normalizes opaque message, agent, workflow-run, and session IDs across the SDK result and notification stream plus the parent and two child JSONL logs. Trusted pull requests add a real-provider two-turn file write/read whose external bytes, tool calls, completed reasons, and persisted log must agree. This harness stays separate from ACP's `pnpm run test:snapshot` because the protocols and build artifacts differ.
The verification surface has three tiers. Mechanism tier: the measured conclusions for the `--sea` chain are embedded in the Decision sections (ESM dynamic import inside the VFS, single cordis instance, fail-loud config chain, `node:sqlite`, macOS ad-hoc signing runs). SDK tier: the complete keyless pytest suite covers the client protocol against a fake runtime peer, subprocess cleanup, absolute cwd propagation, dual-carrier launch, and carrier resolution; root CI runs it on Python 3.10. End-to-end tier: every platform build installs both wheels into a clean venv outside the checkout, proves matching versions and installed module/executable locations, then completes turns against a mock endpoint through the default SDK path, a custom config, the checked-in standalone minimal composition, and the direct binary protocol, with final text and JSONL checked. The minimal run asserts its exact system prompt and single shell-tool catalog and retains shell state across calls. The custom config additionally drives `run_code` and a zero-agent `workflow` through the shared packaged Node process bootstrap. The filesystem-search scenario requires the model to call both `glob` and `grep` through the target-native `-rg` sidecar. The spawn-node scenario drives the platform shell tool through a command starting with `node` and requires the machine's own Node version in the tool result with no `PKG_EXECPATH` in the child environment, pinning the packaged runtime against a pkg upgrade that re-records the child-process patch. The MCP scenario starts a temporary external stdio server, deliberately delays its initial `tools/list` response, then immediately starts the first SDK prompt; the prompt must see and call the discovered tool, proving that `initialize` is a real Loader-settlement readiness boundary rather than a timing sleep. The same installed run compares a committed executable-specific snapshot through the Python SDK: a keyless scripted model mounts a Cordis plugin that registers a tool, invokes that tool from `run_code`, runs a direct spawn subagent and a workflow that starts a second spawn child, then unmounts the plugin. The fixture explicitly disables its unused bundled Bash and local skill discovery so its tool set does not depend on repository-external state, and the comparison normalizes opaque message, agent, workflow-run, and session IDs across the SDK result and notification stream plus the parent and two child JSONL logs. Trusted pull requests add a real-provider two-turn file write/read whose external bytes, tool calls, completed reasons, and persisted log must agree. This harness stays separate from ACP's `pnpm run test:snapshot` because the protocols and build artifacts differ.
Manual-driving caveat: the bin treats stdin EOF as "the client is gone" and disposes immediately, so a short-lived pipe aborts an in-flight turn — pipe-driven runs must keep stdin open until the turn ends.
@@ -40,11 +40,11 @@ exe 的 VFS 内是**构建产物形态的真实包树**(各包的 `lib/` + 真
部署根目录是 [`python/sdk-runtime/package.json`](../../../../python/sdk-runtime/package.json)(`dsh-python-runtime-closure`,pnpm 工作区成员、零代码纯依赖 manifest),也是「exe 安装哪些插件」与「Python 运行时分发什么」的统一真源。向 exe 添加插件,就是在 manifest 中增加一行依赖后重新打包。[`scripts/verify-runtime-closure.ts`](../../../../scripts/verify-runtime-closure.ts) 读取每个已发布的 `packages/preset/agent-presets/presets/*/agent.cordis.yml`,针对 `python/sdk-runtime/platforms.json` 中的每个目标解析比较 `process.platform` 的 `disabled` 条件,并要求该目标启用的每个工作区插件都通过显式的 `workspace:` 依赖列在运行时根目录。它还遍历该 manifest 覆盖的全部工作区包,要求每个非可选的工作区对等依赖(peer dependency)都显式列出,并报告“preset 或引用包 → 缺失依赖”的完整链路;无法识别的平台条件会保持启用,避免因不支持的表达式遗漏插件。`pnpm run hygiene`、CI 静态检查与 single-exe 构建都会在打包前运行该门禁。部署还会依据各包的 `files` 字段打包,因此 tsdown 拆出的共享分片必须被 `files` 覆盖。
部署根目录显式包含 `@deepseek-ai/dsh-mcp-client`,将其作为自定义配置可用的插件,即使随附 preset 均未挂载该插件。外部配置因此可以连接由用户提供的 stdio 与 Streamable HTTP MCP server 并注册其工具;分发物不包含这些 server,也不将桥接范围扩展到 MCP Resources 和 Prompts。可执行程序与已安装 wheel 包的冒烟测试会启动临时 stdio server,发现其工具,并完成一次由模型请求的调用。
部署根目录显式包含 `@deepseek-ai/dsh-mcp-client`,将其作为自定义配置可用的插件,即使随附 preset 均未挂载该插件。外部配置因此可以连接由用户提供的 stdio 与 Streamable HTTP MCP server 并注册其工具;分发物不包含这些 server。可选资源服务提供 MCP Resources;MCP Prompts 仍不受支持。可执行程序与已安装 wheel 包的冒烟测试会启动临时 stdio server,发现其工具,并完成一次由模型请求的调用。
### 构建流水线与产物
[`scripts/build-exe-for-python-sdk.ts`](../../../../scripts/build-exe-for-python-sdk.ts):运行时闭包校验 → `pnpm run build` →(清空后)`pnpm --filter dsh-python-runtime-closure deploy --legacy --prod --config.node-linker=hoisted --config.auto-install-peers=false --config.link-workspace-packages=true` **直接写入** `python/sdk-runtime/src/deepseek_harness_runtime/runtime/node/`,其中包含作为载体根目录 `runtime-bootstrap.mjs` 的 [`python/sdk-runtime/runtime-bootstrap.mjs`](../../../../python/sdk-runtime/runtime-bootstrap.mjs) → 恢复 legacy deploy 遗漏的直接工作区包,并拒绝剩余的 manifest 缺口 → 将暂存依赖中的符号链接替换为目标文件内容,删除包管理器的 `.bin` 链接,并在仍有任何符号链接时失败 → 验证已部署的 bootstrap,并注入以该文件为 bin 的 pkg 配置及覆盖动态读取 profile、bundle、前端、preset、原生库与配置文件的 assets → 暂存目标平台的 `node-pty` addon → 每个构建目标调用一次 `pkg --sea` → 将 `deepseek-harness-sdk-runtime-<platform>-<arch>` 写入 `dist-exe/` 并拷回运行时目录。该 bootstrap 由 Python runtime 拥有;普通启动时它调用公开 CLI export,而提供方私有选择会分派到同一个 `@deepseek-ai/dsh-subprocess-local/runner` 核心,不改变 CLI 语法,也不增加另一个可执行文件;[原生 containment 决策](2026-08-28-subprocess-native-containment.zh.md)负责这条私有路径。Linux CI 会在匹配的 manylinux 2.28 容器中重新构建 `pty.node`,因为 legacy deploy 会遗漏这一安装副作用。每个目标都会把对应的原生 `@vscode/ripgrep` 二进制复制到可执行文件旁,作为必需的 `-rg` 伴随文件;pkg 运行时通过 `process.pkg` 选择该伴随文件,普通 Node 执行则直接使用 `@vscode/ripgrep`。macOS 使用对应目标的预构建产物,并额外生成所需的 `-spawn-helper`。四个部署标志都有实测依据:未启用 `inject-workspace-packages` 时必须使用 `--legacy`;`hoisted` 为 pkg 提供稳定的单实例布局,再由显式物化步骤消除符号链接;关闭对等依赖自动安装可防止未声明的对等依赖扩大闭包;`link-workspace-packages` 选择直接工作区依赖。[`pnpm-workspace.yaml`](../../../../pnpm-workspace.yaml) 将传递的 `@deepseek-ai/cosmokit` 与 `@deepseek-ai/schemastery` semver 请求覆盖到固定的 vendor 源码,使 legacy deploy 不会从注册表解析这些未发布名称。
[`scripts/build-exe-for-python-sdk.ts`](../../../../scripts/build-exe-for-python-sdk.ts):运行时闭包校验 → `pnpm run build` →(清空后)`pnpm --filter dsh-python-runtime-closure deploy --legacy --prod --config.allow-unused-patches=true --config.node-linker=hoisted --config.auto-install-peers=false --config.link-workspace-packages=true` **直接写入** `python/sdk-runtime/src/deepseek_harness_runtime/runtime/node/`,其中包含作为载体根目录 `runtime-bootstrap.mjs` 的 [`python/sdk-runtime/runtime-bootstrap.mjs`](../../../../python/sdk-runtime/runtime-bootstrap.mjs) → 恢复 legacy deploy 遗漏的直接工作区包,并拒绝剩余的 manifest 缺口 → 将暂存依赖中的符号链接替换为目标文件内容,删除包管理器的 `.bin` 链接,并在仍有任何符号链接时失败 → 验证已部署的 bootstrap,并注入以该文件为 bin 的 pkg 配置及覆盖动态读取 profile、bundle、前端、preset、原生库与配置文件的 assets → 暂存目标平台的 `node-pty` addon → 每个构建目标调用一次 `pkg --sea` → 将 `deepseek-harness-sdk-runtime-<platform>-<arch>` 写入 `dist-exe/` 并拷回运行时目录。该 bootstrap 由 Python runtime 拥有;普通启动时它调用公开 CLI export,而提供方私有选择会分派到同一个 `@deepseek-ai/dsh-subprocess-local/runner` 核心,不改变 CLI 语法,也不增加另一个可执行文件;[原生 containment 决策](2026-08-28-subprocess-native-containment.zh.md)负责这条私有路径。Linux CI 会在匹配的 manylinux 2.28 容器中重新构建 `pty.node`,因为 legacy deploy 会遗漏这一安装副作用。每个目标都会把对应的原生 `@vscode/ripgrep` 二进制复制到可执行文件旁,作为必需的 `-rg` 伴随文件;pkg 运行时通过 `process.pkg` 选择该伴随文件,普通 Node 执行则直接使用 `@vscode/ripgrep`。macOS 使用对应目标的预构建产物,并额外生成所需的 `-spawn-helper`。部署标志都有实测依据:未启用 `inject-workspace-packages` 时必须使用 `--legacy`;`hoisted` 为 pkg 提供稳定的单实例布局,再由显式物化步骤消除符号链接;关闭对等依赖自动安装可防止未声明的对等依赖扩大闭包;`link-workspace-packages` 选择直接工作区依赖。[`pnpm-workspace.yaml`](../../../../pnpm-workspace.yaml) 将传递的 `@deepseek-ai/cosmokit` 与 `@deepseek-ai/schemastery` semver 请求覆盖到固定的 vendor 源码,使 legacy deploy 不会从注册表解析这些未发布名称。 生产部署允许未使用的工作区补丁,因为开发工具可能不属于运行时闭包;闭包内包的补丁应用失败仍会中止部署。工作区安装继续严格校验未使用的补丁。
CI 使用 [`.github/workflows/build-exe-for-python-sdk.yml`](../../../../.github/workflows/build-exe-for-python-sdk.yml):[安装后 wheel 包验证](../testing/2026-08-23-installed-python-wheel-black-box-ci.zh.md)在拉取请求上运行 Linux/Windows x64,在 master 推送上运行 Linux ARM64 与两种 macOS 架构。[公开发布工作流](../../archived/process/2026-08-11-python-publication-workflow.md)调用它构建全部五个目标;`workflow_dispatch` 仍可选择部分目标。linux-x64、linux-arm64(`ubuntu-24.04-arm`)、macos-arm64、macos-x64(`macos-15-intel`)与 win-x64(`windows-2025`)分别进行原生构建,并在适用平台缓存 `~/.pkg-cache`;macOS 的 ad-hoc 签名由 pkg 处理。每个平台都把发布形态的 SDK wheel 包与运行时 wheel 包安装到 checkout 外的干净 venv,证明包与可执行文件来源,再通过公开 SDK 与直接 NDJSON JSON-RPC 运行完整 keyless 场景。可信拉取请求与 master 推送还会在各自选定的目标上运行真实 DeepSeek 双轮工具冒烟测试;fork 与 Dependabot head 不会获得密钥。Linux 会检查可执行文件和原生 addon 各自的 GLIBC 依赖,并额外运行 manylinux 2.28 冒烟测试;macOS 则检查 runtime、ripgrep 与 PTY helper 的架构,并验证三个载荷的部署目标都符合 wheel 包标签。完整构建五个目标时保留 6 个产物,每个产物只含一个发布文件:平台无关的 SDK wheel 包与 5 个原生运行时 wheel 包;手动选择部分目标时保留 SDK wheel 与所选运行时 wheel。裸 exe 与源码包只作为测试中间输入。[`.gitlab-ci.yml`](../../../../.gitlab-ci.yml) 只接受版本与根目录 `package.json` 匹配的 `python-v<repository-version>` 标签流水线,构建一个 SDK wheel 包和 5 个原生运行时 wheel 包,再由单个串行任务校验并将这 6 个文件发布到项目的 PyPI 注册表。[`python/sdk-runtime` README](../../../../python/sdk-runtime/README.zh.md)负责 Windows 目标及对 Windows arm64 的明确排除。
@@ -60,14 +60,13 @@ Python 客户端使用所选 profile(默认 `sdk`)、有序 patch 文件和
`dsh-python-runtime-closure` 是私有部署 manifest,`deepseek-harness-sdk-runtime-<platform>-<arch>` 是可执行文件族。协议字段 `serverInfo.name` 是 `deepseek-harness-sdk-runtime`;Python 分发包名是 `deepseek-harness-sdk` / `deepseek-harness-runtime-bin`,导入模块名是 `deepseek_harness` / `deepseek_harness_runtime`。
## 工作线程插件
## 打包后的工作流与代码执行
exe 内支持 `dsh-workflow-worker-thread` 与 `dsh-code-runtime-worker-thread`。两个后端构建后的宿主都通过 `fileURLToPath()` 转换相邻 `lib/worker.cjs` 的 URL,再将所得文件系统字符串传给 `Worker`;pkg 的 Worker 钩子可以用这种形式解析 VFS 内文件。该钩子会把 VFS 内的工作线程文件作为 CommonJS 编译,所以工作线程入口采用 CommonJS。工作流引擎在未构建的源码执行中仍保留 `data:` URL 引导程序,只有构建后的相邻入口使用文件系统字符串。自定义配置的可执行文件冒烟测试会加载两个后端,实际调用 `run_code` 与不启动 agent(智能体)的 `workflow`,并要求两个工作线程都从 pkg 的 VFS 内返回 `42`。
`dsh-workflow-ptc` 通过 `dsh-ptc-runtime-node` 发送自包含 guest 程序。共享提供方通过可执行文件的私有 Node bootstrap 分派启动受管子进程,并使用专用控制通道;不需要独立的工作流 worker 资源。[沙箱 Node 决策](2026-09-11-sandboxed-node-ptc-runtime.zh.md)负责进程与策略生命周期,[工作流沙箱复用](2026-09-13-workflow-ptc-sandbox-reuse.zh.md)负责工作流适配器。
## 测试
验证面分三层。机制层:`--sea` 链路的实测结论内嵌在「决策」各节(VFS 内 ESM 动态 `import()`、单一 Cordis 实例、明确报错的配置链路、`node:sqlite`、macOS ad-hoc 签名可运行)。SDK 层:完整的无密钥 pytest 套件以 mock 运行时对端覆盖客户端协议、子进程清理、绝对 `cwd` 传递、双载体启动与载体解析;根 CI 在 Python 3.10 上运行全部用例。端到端层:每个平台构建都会把两个 wheel 包安装进 checkout 外的干净 venv,证明版本相同以及已安装模块/可执行文件的位置,再通过默认 SDK 路径、自定义配置、仓库内置的独立 minimal 组合和直接二进制协议,对 mock 端点完成轮次,并校验最终文本与 JSONL。minimal 运行会断言其精确系统提示词与双工具目录,跨调用保留 Bash 状态,并调用编辑器。自定义配置还会通过打包进 VFS 的真实工作线程文件执行 `run_code` 和不启动 agent 的 `workflow`。文件系统搜索场景要求模型通过目标平台的 `-rg` 伴随文件调用 `glob` 与 `grep`。spawn-node 场景驱动平台 shell 工具执行以 `node` 开头的命令,要求工具结果给出机器自身的 Node 版本且子进程环境中无 `PKG_EXECPATH`,把打包运行时钉死在「pkg 升级重录 child_process 补丁也不得回归」的行为上。MCP 场景会启动临时外部 stdio server,刻意延迟首次 `tools/list` 响应,随后立即启动第一个 SDK 提示词;该提示词必须看到并调用已发现的工具,从而证明 `initialize` 是真正以 Loader 插件树完全稳定为准的就绪边界,而不是依赖定时 sleep。同一项安装后运行还会经 Python SDK 比较一组检入的 exe 专用快照:无密钥脚本化模型挂载一个会注册工具的 Cordis 插件,从 `run_code` 调用该工具,运行一个直接 spawn 的 subagent 和一个会通过 spawn 启动第二个 subagent 的工作流,随后卸载该插件。该 fixture(测试前置数据)会显式禁用组合包中未使用的 Bash 和本地 skill(技能)发现,使其工具集不依赖仓库外部状态;比较时会规范化 SDK 结果与通知流,以及父会话和两个子会话 JSONL 日志中不透明的消息、agent、工作流运行与会话 ID。可信拉取请求会增加真实提供方双轮文件写入/读取,并要求外部字节、工具调用、已完成原因与持久化日志一致。该 harness 与 ACP 的 `pnpm run test:snapshot` 保持独立,因为二者的协议和构建产物不同。
验证面分三层。机制层:`--sea` 链路的实测结论内嵌在「决策」各节(VFS 内 ESM 动态 `import()`、单一 Cordis 实例、明确报错的配置链路、`node:sqlite`、macOS ad-hoc 签名可运行)。SDK 层:完整的无密钥 pytest 套件以 mock 运行时对端覆盖客户端协议、子进程清理、绝对 `cwd` 传递、双载体启动与载体解析;根 CI 在 Python 3.10 上运行全部用例。端到端层:每个平台构建都会把两个 wheel 包安装进 checkout 外的干净 venv,证明版本相同以及已安装模块/可执行文件的位置,再通过默认 SDK 路径、自定义配置、仓库内置的独立 minimal 组合和直接二进制协议,对 mock 端点完成轮次,并校验最终文本与 JSONL。minimal 运行会断言其精确系统提示词与单个 shell 工具目录,并跨调用保留 shell 状态。自定义配置还会通过共享的打包后 Node 进程 bootstrap 执行 `run_code` 和不启动 agent 的 `workflow`。文件系统搜索场景要求模型通过目标平台的 `-rg` 伴随文件调用 `glob` 与 `grep`。spawn-node 场景驱动平台 shell 工具执行以 `node` 开头的命令,要求工具结果给出机器自身的 Node 版本且子进程环境中无 `PKG_EXECPATH`,把打包运行时钉死在「pkg 升级重录 child_process 补丁也不得回归」的行为上。MCP 场景会启动临时外部 stdio server,刻意延迟首次 `tools/list` 响应,随后立即启动第一个 SDK 提示词;该提示词必须看到并调用已发现的工具,从而证明 `initialize` 是真正以 Loader 插件树完全稳定为准的就绪边界,而不是依赖定时 sleep。同一项安装后运行还会经 Python SDK 比较一组检入的 exe 专用快照:无密钥脚本化模型挂载一个会注册工具的 Cordis 插件,从 `run_code` 调用该工具,运行一个直接 spawn 的 subagent 和一个会通过 spawn 启动第二个 subagent 的工作流,随后卸载该插件。该 fixture(测试前置数据)会显式禁用组合包中未使用的 Bash 和本地 skill(技能)发现,使其工具集不依赖仓库外部状态;比较时会规范化 SDK 结果与通知流,以及父会话和两个子会话 JSONL 日志中不透明的消息、agent、工作流运行与会话 ID。可信拉取请求会增加真实提供方双轮文件写入/读取,并要求外部字节、工具调用、已完成原因与持久化日志一致。该 harness 与 ACP 的 `pnpm run test:snapshot` 保持独立,因为二者的协议和构建产物不同。
手工驱动注意:`bin` 将 stdin EOF 视为「客户端已离开」并立即 dispose,生命周期较短的管道会中止进行中的轮次——管道驱动必须保持 stdin 打开,直到轮次结束。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-12-agent-scope-runtime-design.md
2026-07-12-agent-scope-runtime-design.md: ca300d4eeeab878a4e41b8e68a669be418617181
2026-07-12-agent-scope-runtime-design.zh.md: 870690d6ace9fefd859557a2e73e88b9b1da6206
2026-07-12-agent-scope-runtime-design.md: b9b2c837f560c2b8b98da37e3656caf298b705e2
2026-07-12-agent-scope-runtime-design.zh.md: 4b94d1b8449bd619c47f6bee0712ec8ed0b39593
@@ -60,6 +60,8 @@ Product helpers therefore construct the carrier and pass the domain subject sepa
A Cordis waterfall is middleware-style dispatch. Each listener receives `next()`: calling it delegates to the remaining listeners and base operation, while returning without it short-circuits or replaces the downstream result. Waterfalls power prompt assembly and tool policy; ordinary emit events notify synchronously, and parallel events await all listeners without a veto result.
<a id="scope-routing-one-opaque-key-selects-one-layer"></a>
## Scope routing: one opaque key selects one layer
The scope package implements the smallest object needed for Cordis routing. Its carrier holds only a composed service filter and scope predicate, while the package records the opaque key privately and exposes the scope fiber's quiescent disposer separately.
@@ -117,30 +119,28 @@ Publication admits and announces resources in the order required by observers:
1. Enter the session.
2. Enter the agent.
3. Announce `session/created`.
4. Announce `agent/created`.
5. Enable public driving.
6. Emit `agent/session-start`.
7. Start the driver.
4. Await serial `agent/created` listeners.
5. Release queued input to the driver.
The agent never drives before both registries and creation notifications agree. A synchronous listener may veto or dispose an owner; the transaction records publication in progress and waits for that callback stack to unwind before teardown continues. Every creation announcement that begins has a matching disposal announcement during rollback.
The agent never drives before both registries and creation listeners finish. A listener may reject or dispose an owner; the transaction retains the scope and session until dispatch settles before teardown continues. Every creation announcement that begins has a matching disposal announcement during rollback. The [awaited creation decision](2026-09-09-awaited-agent-creation.md) owns asynchronous initializer timing.
The sequence diagram isolates the non-obvious race: a synchronous creation listener can request disposal while the publication call stack still owns both registry entries. Teardown must deactivate immediately but wait for that stack to unwind before stopping and detaching anything.
A creation listener can request disposal while publication still owns both registry entries. Teardown deactivates immediately and waits for the awaited dispatch before stopping and detaching anything.
```mermaid
sequenceDiagram
participant Tx as AgentCreationTransaction
participant Registries
participant Listener as Synchronous listener
participant Listener as Creation listener
participant Driver
Tx->>Tx: mark publication in progress
Tx->>Registries: announce agent/created
Registries->>Listener: invoke inside the same call stack
Registries->>Listener: await listener
Listener->>Tx: dispose reentrantly
Tx->>Tx: deactivate, teardown waits for publication
Tx-->>Listener: disposal request accepted
Listener-->>Registries: return
Registries-->>Tx: announcement unwound
Listener-->>Registries: settle
Registries-->>Tx: dispatch settled
Tx->>Tx: resolve publication settlement
Tx->>Driver: stop and drain
Tx->>Registries: detach agent, then session
@@ -151,7 +151,7 @@ sequenceDiagram
Every teardown request joins one memoized path. The order is:
1. Deactivate creation or driving and let synchronous publication finish.
1. Deactivate creation or driving and await creation dispatch.
2. Stop and drain the driver, discarding any injection that remains pending.
3. Detach the agent.
4. Detach the session.
@@ -160,6 +160,8 @@ Every teardown request joins one memoized path. The order is:
This order lets final agent and session events use the matching scoped listeners and keeps persistence observers attached through the final flush. Scope disposal comes last because registration revocation is the externally visible lifetime boundary.
<a id="session-append-materialize-validate-commit-notify"></a>
## Session append: materialize, validate, commit, notify
Session events cross a durable boundary, so append owns their data. The rest of the algorithm uses one attached entry and one commit point.
@@ -232,6 +234,8 @@ This is a trusted same-process extension point, not an authority boundary. A lis
Scope solves the real isolation problem directly. Structured-output contributions register in the child's exact scope, while PTC mode derives its transport and SDK from the same resolved tool view. A second named-protection system would need another ownership and collision rule across arbitrary schema providers—including providers that intentionally contribute duplicate names—without creating a new trust boundary.
<a id="structured-output-commits-only-authoritative-outcomes"></a>
### Structured output commits only authoritative outcomes
Structured output combines child-scoped composition with a two-phase execution commit. The child registers its `structured_output` tool and instruction before publication; a trusted assembly listener may transform those ordinary contributions and is responsible for preserving the protocol if the child is expected to complete. The tool body validates a candidate and stages it by the current `ToolExecution`, but successful capture is decided only by immutable `tools/result` observations.
@@ -244,6 +248,8 @@ Once a value is pending or committed, a scoped monotonic guard denies later tool
Pure PTC mode's registry contribution omits `structured_output` from native wire schemas and exposes it through the generated SDK. The assembly waterfall may deliberately change that presentation; execution still validates against the child-scoped definition, and the listener owns the consistency of any alternate model-visible route it creates.
<a id="three-execution-boundaries-are-deliberately-one-way"></a>
### Three execution boundaries are deliberately one-way
Prompt assembly is intentionally cooperative, but three execution facts need one-way settlement after their extensible stages:
@@ -294,19 +300,21 @@ Start resolves only after `initialize` and `newSession` succeed. Abort, spawn fa
Worker and child-process bridges need more state than same-process registries because messages, process death, and cleanup can settle independently. Their state is organized around those real facts rather than duplicate cancellation protocols.
<a id="workflow-children-are-pending-starts-or-published-records"></a>
### Workflow children are pending starts or published records
The workflow host keeps pending provider-start promises and published child records. A child moves from pending to published only when async `SubagentRuntime.start()` fulfills; rejected starts clean their partial provider work and produce no child lifecycle pair.
One host-owned AbortController supplies the required signal to pending and live children. Closing workflow admission aborts that signal, so there is no duplicate `ChildCancel` worker RPC or explicit host-side `run.cancel()` fanout. Quiescence waits for both pending starts and published child disposal.
One host-owned AbortController supplies the required signal to pending and live children. Closing workflow admission aborts that signal; quiescence waits for both pending starts and published child disposal. [Workflow sandbox reuse](2026-09-13-workflow-ptc-sandbox-reuse.md) owns PTC process cancellation and the absence of a separate workflow cleanup timer.
The worker boundary still serializes requests and outcomes. The host retains first-terminal-outcome arbitration, exact child accounting, worker-death handling, grace termination, late/duplicate message rejection, and bounded cleanup because result receipt, worker exit, and child quiescence are genuinely independent facts.
PTC serializes requests and outcomes and owns process termination. The workflow adapter retains terminal-outcome arbitration and child ownership because program settlement, process exit and child quiescence remain independent facts.
### Terminal result and physical cleanup remain separate
The workflow result records the first accepted terminal outcome according to the public precedence rules. Cleanup can continue after that result is chosen: live children still need disposal, a worker still needs termination, and a slow external backend may outlive the configured grace bound.
The workflow result records the first accepted terminal outcome according to the public precedence rules. Choosing that outcome does not release resources: the PTC process and live children still need cleanup, and child disposal must fulfill its provider contract.
Public disposal claims its memoized promise before invoking callbacks. Worker death closes admission before processing any queued late child request, synthesizes missing lifecycle ends, and starts child/process cleanup without rewriting an outcome already claimed.
Public disposal joins one cleanup operation. Run settlement closes child admission, synthesizes missing lifecycle ends and cleans up children without rewriting an outcome already claimed.
### ACP prompt settlement does not depend on update delivery
@@ -334,7 +342,7 @@ The plugin does not police trusted setup by scanning registries or reject prompt
The event catalog, service catalog, producer/consumer matrix, configuration catalog, module graph, tool catalog, type-equivalence blocks, and scoped-event resolver map are generated or freshness-gated from source. The [TypeScript semantic-gates Agent Note](../../archived/process/2026-07-14-typescript-program-backed-semantic-gates.md) owns Program construction, semantic event discovery, and resolver-generation rules.
Behavioral tests pin scoped routing and disposal, final-entry collision cleanup, publication rollback, ordered quiescence, durable pre/post-commit behavior, live tool filtering across presentation and execution, cooperative prompt assembly, structured-output commit in native and PTC mode, async subagent startup and signal cancellation, worker terminal arbitration, ACP settlement, and process teardown.
Behavioral tests pin scoped routing and disposal, final-entry collision cleanup, publication rollback, ordered quiescence, durable pre/post-commit behavior, live tool filtering across presentation and execution, cooperative prompt assembly, structured-output commit in native and PTC mode, async subagent startup and signal cancellation, workflow terminal arbitration, ACP settlement, and process teardown.
## Alternatives considered
@@ -119,30 +119,28 @@ Setup 接收完整的子上下文和确切的未发布 Agent,可以等待插
1. 将会话写入注册表。
2. 将 agent 写入注册表。
3. 宣告 `session/created`。
4. 宣告 `agent/created`。
5. 启用公开驱动。
6. 发射 `agent/session-start`。
7. 启动 driver。
4. 等待串行 `agent/created` 监听器。
5. 向驱动器释放已排队输入。
Agent 在两个注册表和创建通知都达成一致之前绝不驱动。同步监听器可以否决或 dispose 一个所有者;事务记录发布进行中,并等待该回调栈展开后再继续拆除。每个已开始的创建宣告在回滚期间都有匹配的销毁宣告。
Agent 在两个注册表与创建监听器都完成前绝不驱动。监听器可以拒绝或 dispose 一个所有者;事务保留作用域与会话,等待分发结算后再继续拆除。每个已开始的创建宣告在回滚期间都有匹配的销毁宣告。[可等待创建决策](2026-09-09-awaited-agent-creation.zh.md)拥有异步初始化器时序。
以下序列图隔离了非显而易见的竞态:同步创建监听器可以在发布调用栈仍拥有两个注册表条目时请求 dispose。拆除必须立即停用,但要等待该栈展开后才停止和分离任何东西。
创建监听器可以在发布仍拥有两个注册表条目时请求 dispose。Teardown 会立即停用,并等待所调用的异步分发完成后才停止和分离资源。
```mermaid
sequenceDiagram
participant Tx as AgentCreationTransaction
participant Registries
participant Listener as Synchronous listener
participant Listener as Creation listener
participant Driver
Tx->>Tx: mark publication in progress
Tx->>Registries: announce agent/created
Registries->>Listener: invoke inside the same call stack
Registries->>Listener: await listener
Listener->>Tx: dispose reentrantly
Tx->>Tx: deactivate, teardown waits for publication
Tx-->>Listener: disposal request accepted
Listener-->>Registries: return
Registries-->>Tx: announcement unwound
Listener-->>Registries: settle
Registries-->>Tx: dispatch settled
Tx->>Tx: resolve publication settlement
Tx->>Driver: stop and drain
Tx->>Registries: detach agent, then session
@@ -153,7 +151,7 @@ sequenceDiagram
每个拆除请求加入一条记忆化路径。顺序为:
1. 停用创建或驱动,让同步发布完成。
1. 停用创建或驱动,并等待创建分发。
2. 停止并排空 driver,丢弃仍处于待处理状态的注入。
3. 分离 agent。
4. 分离会话。
@@ -308,15 +306,15 @@ Worker 和子进程桥接比同进程注册表需要更多状态,因为消息
工作流宿主保持待定的提供方 start promise 和已发布的子级记录。子级仅在异步 `SubagentRuntime.start()` 兑现时才从待定变为已发布;被拒绝的 start 清理其部分提供方工作且不产生子级生命周期对。
一个宿主拥有的 AbortController 向待定和活跃子级提供必需的 signal。关闭工作流准入中止该 signal,因此没有重复的 `ChildCancel` worker RPC 或显式的宿主侧 `run.cancel()` 扇出。完全停稳需要等待待定 start 和已发布子级 dispose 两者。
一个宿主拥有的 AbortController 向待定和活跃子级提供必需的 signal。关闭工作流准入中止该 signal;完全停稳需要等待待定 start 和已发布子级 dispose 两者。[工作流沙箱复用](2026-09-13-workflow-ptc-sandbox-reuse.zh.md)负责 PTC 进程取消和不另设工作流清理定时器的规则。
Worker 边界仍然序列化请求和结果。宿主保留首个终端结果仲裁、精确的子级计数、worker 死亡处理、优雅终止、迟到/重复消息拒绝和有界清理,因为结果接收、worker 退出和子级完全停稳是真正独立的事实。
PTC 序列化请求和结果并负责进程终止。工作流适配器保留终态结果仲裁和子级归属,因为程序结算、进程退出和子级完全停稳仍是独立事实。
### 终端结果与物理清理保持分离
工作流结果按公开优先级规则记录首个被接受的终端结果。该结果选定后清理可以继续:活跃子级仍需 dispose,worker 仍需终止,慢速外部后端可能超出配置的优雅期限。
工作流结果按公开优先级规则记录首个被接受的终端结果。选定结果不会释放资源:PTC 进程和活跃子级仍需清理,子级资源释放必须履行其提供方约定。
公开 dispose 在调用回调之前取得其记忆化 promise 的所有权。Worker 死亡在处理任何排队的迟到子级请求之前关闭准入,合成缺失的生命周期结束,并启动子级/进程清理而不重写已声明的结果。
公开 dispose 汇入同一个清理操作。运行结算关闭子级准入,合成缺失的生命周期结束并清理子级,不重写已经认领的结果。
### ACP 提示词结算不依赖更新投递
@@ -344,7 +342,7 @@ TypeScript 无法管控 JavaScript 强制转换、直接 Cordis dispatch、进
事件目录、服务目录、生产者/消费方矩阵、配置目录、模块图、工具目录、type-equiv 块和作用域事件解析器映射都是从源码生成或受新鲜度门禁约束的。[TypeScript 语义门禁 Agent Note](../../archived/process/2026-07-14-typescript-program-backed-semantic-gates.md) 拥有 Program 构造、语义事件发现和解析器生成规则。
行为测试固定了作用域路由和 dispose、最终写入注册表时的碰撞清理、发布回滚、有序完全停稳、持久化前/后提交行为、跨展示和执行的活跃工具过滤、协作式提示词组装、原生和 PTC mode 中的结构化输出提交、异步 subagent 启动和信号取消、worker 终端仲裁、ACP 结算和进程拆除。
行为测试固定了作用域路由和 dispose、最终写入注册表时的碰撞清理、发布回滚、有序完全停稳、持久化前/后提交行为、跨展示和执行的活跃工具过滤、协作式提示词组装、原生和 PTC mode 中的结构化输出提交、异步 subagent 启动和信号取消、工作流终态仲裁、ACP 结算和进程拆除。
## 曾考虑的替代方案
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-14-provider-routed-llm-adapters.md
2026-07-14-provider-routed-llm-adapters.md: 4cc3cc3cdceeefdfea864ccf3465e528e32159c7
2026-07-14-provider-routed-llm-adapters.zh.md: 098a6af7c951c6dee8a1b4f948bc69725e8d189c
2026-07-14-provider-routed-llm-adapters.md: 1001b10e18837399b41578658ffe62065e294ba8
2026-07-14-provider-routed-llm-adapters.zh.md: c5bff0ef407c8a2c22e8cad7e7d32c58c8b0e2da
@@ -30,7 +30,7 @@ A provider has exactly one adapter owner in a Cordis context. `dsh-llm-deepseek`
`dsh-llm-pi-ai` takes one non-empty list of provider profiles. Provider names must be unique within the list and present in pi-ai's `getProviders()` result. Each profile contains the provider name plus optional `apiKey`, `baseURL`, headers, reasoning level and budgets, cache retention, transport, SDK timeouts, a Harness stream-idle timeout, and a provider-owned `retryPolicy`. The adapter forces pi-ai's `maxRetries` to zero so one `stream()` call makes one visible provider attempt, while `dsh-llm-retry` executes the resolved policy at the agent failed-step extension point. Credentials are never global: an explicit key applies only to its profile, while an absent key lets pi-ai resolve its standard environment variable, OAuth token, AWS credential chain, Google ADC, or other provider-native ambient authentication. An explicitly empty key is invalid configuration rather than an environment fallback.
The plugin registers all configured provider names against one `PiAiAdapter` in one all-or-nothing call. A request uses its provider to select the matching profile and finds its model in `getModels(provider)` to obtain the catalog descriptor. An unknown provider fails at plugin load; an unknown model fails before network I/O with `UNKNOWN_MODEL`. The catalog object is never mutated. When a profile supplies `baseURL`, the adapter clones the selected descriptor and overrides only `baseUrl`, so a private endpoint can retain pi-ai's API, capabilities, compatibility flags, context limits, and reasoning map. The private endpoint must implement the selected provider's protocol, and the model id must still exist in the installed pi-ai catalog.
The plugin registers configured provider names against one `PiAiAdapter` in one atomic call. Each immutable request snapshot combines the effective profiles and their serviceable model descriptors. Catalog-external models require an explicit or inferable protocol and endpoint. Stored catalog errors remain visible and repairable under the [settings catalog recovery decision](../bug-fix/2026-09-07-pi-ai-settings-catalog-recovery.md), while writes validate changed providers and requests reject the selected failed model before network I/O.
The adapter calls pi-ai's `streamSimple()` so each catalog model chooses its registered API implementation, including OpenAI Responses instead of Chat Completions where the descriptor says `openai-responses`. Harness temperature, maximum tokens, signal, session id, and the profile's common stream options flow through directly. Profile headers merge with the mandatory Harness attribution headers, with Harness attribution winning its reserved names. The adapter no longer maintains DeepSeek-specific payload rewrites or a provider-protocol matrix.
@@ -54,7 +54,7 @@ Compaction configuration gains `summarizationProvider` beside `summarizationMode
The JSON-RPC runtime receives provider and model explicitly. Its convenience fallback mounts `dsh-llm-deepseek` only for provider `deepseek` when that provider has no registered owner; other missing providers fail without guessing an adapter.
Current v1 seed/load validation rejects request headers and assistant messages that omit required provider/model fields. The frozen v0-to-v1 edge requires the same reconstructable routing identity before migration; it never guesses a missing provider or model, and malformed shapes refuse before publication.
Current seed/load validation rejects request headers and assistant messages that omit required provider/model fields. The frozen v0-to-v1 edge requires the same reconstructable routing identity before migration; it never guesses a missing provider or model, and malformed shapes refuse before publication.
## Alternatives considered
@@ -78,7 +78,7 @@ Current v1 seed/load validation rejects request headers and assistant messages t
- pi-ai credentials, transport knobs, SDK timeouts, and the five-minute-default `streamIdleTimeoutMs` watchdog are scoped per provider profile. Hidden provider retries are disabled; bounded retries belong to the separately composed agent recovery policy.
- `dsh-llm-pi-ai` rejects stop sequences because pi-ai's common stream API cannot express them; the native DeepSeek adapter retains its stop support.
- Replay state is portable only within the adapter instance that owns both the historical and target providers. Cross-provider and cross-model restoration is an adapter responsibility, and another adapter receives provider-neutral history without the opaque state.
- Current v1 Session JSONL requires provider/model on request headers and assistant messages. The v0 edge migrates only frozen shapes that already carry reconstructable request identity.
- Current Session JSONL requires provider/model on request headers and assistant messages. The v0 edge migrates only frozen shapes that already carry reconstructable request identity.
## Testing
@@ -30,7 +30,7 @@ Status: implemented
`dsh-llm-pi-ai` 接受一个非空的提供方配置列表。列表内的提供方名称必须唯一,并且存在于 pi-ai 的 `getProviders()` 结果中。每项配置包含提供方名称,以及可选的 `apiKey`、`baseURL`、headers、推理级别和预算、缓存保留设置、传输方式、SDK 超时、Harness 流空闲超时,以及由提供方拥有的 `retryPolicy`。适配器强制将 pi-ai 的 `maxRetries` 设为零,使一次 `stream()` 调用只发起一次可见的提供方请求;`dsh-llm-retry` 则在 agent 失败步骤扩展点上执行解析后的策略。凭据不设全局值:显式密钥仅对所属配置生效;未提供密钥时,pi-ai 使用标准环境变量、OAuth token、AWS 凭据链、Google ADC 或其他提供方原生环境认证。显式空密钥属于无效配置,不会回退到环境认证。
插件通过一次全有或全无调用,将所有已配置的提供方名称注册到同一个 `PiAiAdapter`。请求按 provider 选择对应配置,并在 `getModels(provider)` 中查找模型以取得目录描述符。未知提供方会在插件加载时失败;未知模型会在网络 I/O 前以 `UNKNOWN_MODEL` 失败。适配器不会修改目录对象。当配置提供 `baseURL` 时,适配器复制选中的描述符,仅覆盖 `baseUrl`,使私有端点保留 pi-ai 的 API、能力、兼容标志、上下文限制与推理映射。私有端点必须实现所选提供方的协议,模型 ID 也仍须存在于已安装的 pi-ai 目录中。
插件通过一次原子调用,将已配置的提供方名称注册到同一个 `PiAiAdapter`。每个不可变请求快照组合有效 profile 与可服务模型的描述符。目录外模型需要显式指定或可推断的协议与端点。根据[设置目录恢复决策](../bug-fix/2026-09-07-pi-ai-settings-catalog-recovery.zh.md),已存储的目录错误保持可见、可修复,写入仍会校验已修改提供方,请求则在网络 I/O 前拒绝所选错误模型。
适配器调用 pi-ai 的 `streamSimple()`,因此每个目录模型会选择其注册的 API 实现;描述符为 `openai-responses` 时使用 OpenAI Responses,而非 Chat Completions。Harness 的 temperature、最大 token 数、signal、session ID,以及提供方配置中的通用流选项均直接传递。配置 headers 与 Harness 强制归因 headers 合并;发生保留名称冲突时,以 Harness 归因为准。适配器不再维护 DeepSeek 专用 payload 重写或提供方协议矩阵。
@@ -54,7 +54,7 @@ pi-ai 回放状态用其成功 `AssistantMessage` 的带版本最小投影填充
JSON-RPC 运行时显式接收提供方与模型。仅当 `deepseek` 提供方没有注册所有者时,其便利回退才会挂载 `dsh-llm-deepseek`;其他缺失的提供方会直接失败,不会猜测适配器。
当前 v1 的 seed/load 验证会拒绝省略必需提供方/模型字段的请求头和助手消息。冻结的 v0-to-v1 迁移边要求迁移前已具备同一套可重建路由身份;它绝不会猜测缺失的提供方或模型,畸形结构会在发布前被拒绝。
当前的 seed/load 验证会拒绝省略必需提供方/模型字段的请求头和助手消息。冻结的 v0-to-v1 迁移边要求迁移前已具备同一套可重建路由身份;它绝不会猜测缺失的提供方或模型,畸形结构会在发布前被拒绝。
## 考虑过的替代方案
@@ -78,7 +78,7 @@ JSON-RPC 运行时显式接收提供方与模型。仅当 `deepseek` 提供方
- pi-ai 凭据、传输选项、SDK 超时,以及默认五分钟的 `streamIdleTimeoutMs` 空闲超时机制均按提供方配置隔离。系统禁用隐藏的提供方重试;有界重试由单独组合的 agent 恢复策略负责。
- pi-ai 的通用流 API 无法表达停止序列,因此 `dsh-llm-pi-ai` 会拒绝停止序列;原生 DeepSeek 适配器仍支持停止序列。
- 仅当历史提供方与目标提供方归同一个适配器实例所有时,回放状态才可移植。适配器负责跨提供方和跨模型恢复;其他适配器只接收不含不透明状态的提供方无关历史。
- 当前 v1 Session JSONL 要求请求头和助手消息都包含提供方/模型。v0 边只迁移已经携带可重建请求身份的冻结结构。
- 当前 Session JSONL 要求请求头和助手消息都包含提供方/模型。v0 边只迁移已经携带可重建请求身份的冻结结构。
## 测试
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-22-slot-type-chain-implementation.md
2026-07-22-slot-type-chain-implementation.md: 0e493a19e232acf6f289714e53e0cf64b3ffc4b0
2026-07-22-slot-type-chain-implementation.zh.md: 5e5c2ec808cec6949396a333c6a86e267bb4268b
2026-07-22-slot-type-chain-implementation.md: 98585a42f71844594da4ada19d7438a7d8321702
2026-07-22-slot-type-chain-implementation.zh.md: d1093537ccbc5ee5ff2dc18c234233ffca39879d
@@ -12,6 +12,8 @@ The page is composed at runtime from independently loaded plugins, so the UI nee
## Decision
Global main-panel selection and its root lifetime are defined by the [global main-panels decision](2026-09-08-global-main-panels.md).
One sentence: **the ui-renderer renders only `'root'`; a plugin composes UI through a single `register` call that simultaneously occupies a slot, declares+authorizes its child slots, declares its store, and injects its business face; components are pure functions whose props arrive in four shares, each auto-derived from its single source of truth.**
### 'root' is the only a-priori slot
@@ -25,7 +27,7 @@ ctx.slots.register({
name: 'root',
children: {
'sidebar': { kind: 'single', scope: 'root' },
'conversation': { kind: 'single', scope: 'session' },
'main': { kind: 'keyed', scope: 'root' },
},
store: createLayoutStore, // StoreHandle or factory (below)
inject: injectFrame, // business face (below)
@@ -12,6 +12,8 @@ Status: implemented
## 决策
全局主面板选择及其 root 生命周期由[全局主面板决策](2026-09-08-global-main-panels.zh.md)定义。
一句话:**ui-renderer 只渲染 `'root'`;插件用单独一次 `register` 调用组合 UI——这一次调用同时占用 slot、声明并授权子 slot、声明 store、注入业务面;组件是纯函数,props 分四份额到达,每一份额都从各自唯一的真源自动推导。**
### 'root' 是唯一的先验 slot
@@ -25,7 +27,7 @@ ctx.slots.register({
name: 'root',
children: {
'sidebar': { kind: 'single', scope: 'root' },
'conversation': { kind: 'single', scope: 'session' },
'main': { kind: 'keyed', scope: 'root' },
},
store: createLayoutStore, // StoreHandle or factory (below)
inject: injectFrame, // business face (below)
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-23-client-plugin-loading-model.md
2026-07-23-client-plugin-loading-model.md: 21bad78792c6b5aad48b51f454f6c08c0400ad72
2026-07-23-client-plugin-loading-model.zh.md: 2758f28f3bd34131ece3bed74152fbfe0b36174e
2026-07-23-client-plugin-loading-model.md: d0f9b20f0adabda6cc7132e5411bcadd60c9e108
2026-07-23-client-plugin-loading-model.zh.md: 27d5026ed03509d6408a16389246cb1321a37959
@@ -52,18 +52,18 @@ What happens between `dsh web` starting and the UI appearing? Three stages: the
**Host side — compose the graph.**
1. The composing app (`apps/cli`) ships the roster as ordinary rows in its `cordis.yml` config tree — client plugin packages are entry rows like every host plugin, including the always-mounted `client-hmr` row. A roster row that fails to import is caught by `assertEntriesLoaded`; a row whose fiber rejects is reported with its original stack by `assertEntriesActivated` ([host boot decision](2026-07-24-web-config-tree-boot-and-transport-layering.md)).
1. The composing app (`apps/cli`) ships the roster as ordinary rows in its `cordis.yml` config tree — client plugin packages are entry rows like every host plugin, including the always-mounted `client-hmr` row. `auditStartupEntries` reports failed imports, activation errors with their original stacks, and pending dependencies; optional entries warn and required entries reject startup ([startup policy](2026-09-09-consumer-owned-startup-strictness.md)).
2. The `dsh-client-modules` node half (the package is dual-face: its browser half is the module table) resolves each live Loader entry through the same `name` and owning-tree `baseUrl` inputs that imported its Host face, then reads the nearest owning package.json `dsh.client` declaration and composes `window.__DSH_BOOT__`: `{ rev, entries: [{ id, url, rev, inject?, immediately?, external? }], batches: [{ phase, url, rev, entries }] }`. The manifest package name is the browser module identity even when an overlay names a relative source or built entry file. Distinct active Loader sources resolving to one package name fail composition; after one source unloads, the surviving source supplies the row without a fiber restart. The row's three optional fields come from manifests, never hand-copied. Composition orders requested dynamic rows before their consumers, rejects synchronous request cycles, and assigns every row to exactly one initial batch. It refuses declared plugins without built `./client` bundles and groups their package/path rows under one required source-build instruction; malformed declaration fields also fail activation, and the Host audit reports either error from the FAILED fiber.
3. Scanning is incremental per package — there is no full-rescan code path. Each cordis `internal/plugin` emission marks the fiber's entry name dirty (entry-less fibers drop O(1)); a microtask flush reconciles each dirty name against live loader entries, with package metadata (including the negative "not a client package" verdict) cached per entry name and owning-tree base URL for the process lifetime and bundle re-hashing reachable only through `rebuilt(id)`. The activation pass seeds the same dirty set from current entries and flushes synchronously, so first scan and steady state share one implementation. Initial rows receive an opaque process nonce plus sequence without hashing their artifacts; startup combo revisions hash the combined script inputs plus indexed map, and the rows plus batch descriptors hash into `graph.rev`. The graph types are single-sourced in the modules package's `./client` export — the webserver knows nothing about the graph, while modules registers the combo route and contributes structured index-injection rows.
Why is the roster yml rows and not a scan? Because which plugins compose into a deployment is a composition decision, not a package property — a package declaring `dsh.client` in the repo does not mean this deployment mounts it, so discovery-by-scan cannot make that call; the node half scans only what the tree actually mounted.
**Phase one — the module face.** The injected HTML installs `window.__ModuleLoader__` in queue mode, starts preloading every application combo URL, executes every bootstrap combo URL as a blocking classic script, assigns `window.__DSH_BOOT__`, and then starts the Vite main module. The kernel calls the facade's `create()` with the raw graph and shell seeds. The facade removes and materializes the modules registration with a bootstrap `require` that rejects every external, then calls its `createClientModuleSystem` export. The modules bundle parses the graph, constructs the system, memoizes its own exports, retains the instance in its module closure, and switches the same facade to live registration. The kernel then prefetches every `immediately` row in parallel. Rows in the same application combo share its execution; separate combos load independently when an immediate row, a requested dependency, or ordinary entry import reaches them. A prefetch failure is swallowed here because phase two's import retries and owns the loud failure. `immediately` remains a registration barrier, not a package identity.
**Phase one — the module face.** The injected HTML installs `window.__ModuleLoader__` in queue mode, starts preloading every application combo URL, executes every bootstrap combo URL as a blocking classic script, assigns `window.__DSH_BOOT__`, and then starts the Vite main module. The kernel calls the facade's `create()` with the raw graph and shell seeds. The facade removes and materializes the modules registration with a bootstrap `require` that rejects every external, then calls its `createClientModuleSystem` export. The modules bundle parses the graph, constructs and returns the system, memoizes its own exports, and switches the same facade to live registration. The kernel then prefetches every `immediately` row in parallel. Rows in the same application combo share its execution; separate combos load independently when an immediate row, a requested dependency, or ordinary entry import reaches them. A prefetch failure is swallowed here because phase two's import retries and owns the loud failure. `immediately` remains a registration barrier, not a package identity.
**Phase two — the plugin face.**
1. The kernel mounts the vendored Loader and injects the module system as `internal` before any entry exists. Ordering matters: `tree.import`'s bare-import fallback must never run in a browser.
2. It creates every graph row uniformly. Importing the modules row returns the memoized bootstrap exports, whose `apply()` provides the closed-over system as `ctx.modules`; rows that require that service remain PENDING until then, so the modules row needs no special creation position. Render assembly is an ordinary host-graph row provided by `dsh-client-ui-renderer`; the kernel appends no assembly pseudo-entry.
2. It creates every graph row uniformly. Importing the modules row returns the memoized bootstrap exports, whose `apply()` reads that tree's `Loader.internal` and provides the same instance as `ctx.modules`; rows that require that service remain PENDING until then, so the modules row needs no special creation position. Render assembly is an ordinary host-graph row provided by `dsh-client-ui-renderer`; the kernel appends no assembly pseudo-entry.
3. Graph order governs synchronous factory availability; Cordis activation remains independent and proceeds through service waiting.
4. `settled` = every entry created + `loader.await()` quiescent + an all-ACTIVE sweep. The sweep lists each import-failed, FAILED, or PENDING fiber with its missing services. It exists because cordis inject waits have no timeout — the sweep is the fail-loud floor.
5. The framework-free loading page projects real fiber states via `internal/status`. After the sweep, the kernel calls `ctx.uiRenderer.mount(container)` and replaces the page with the real UI in one pass.
@@ -52,18 +52,18 @@ Host 会快照每个已构建插件产物,并把每个调度阶段的有序 ro
**host 侧——组合这张图。**
1. 负责组合的 app(`apps/cli`)把名册作为普通行放进它的 `cordis.yml` 配置树——client 插件包与每个 host 插件一样是 entry 行,包括无条件挂载的 `client-hmr` 行。名册行 import 失败由 `assertEntriesLoaded` 捕获;fiber reject 的行则由 `assertEntriesActivated` 报告原始 stack([host boot 决策](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md))。
1. 负责组合的 app(`apps/cli`)把名册作为普通行放进它的 `cordis.yml` 配置树——client 插件包与每个 host 插件一样是 entry 行,包括无条件挂载的 `client-hmr` 行。`auditStartupEntries` 报告 import 失败、带原始 stack 的激活错误,以及待满足的依赖;optional entry 输出 warning,required entry 则使启动失败([启动策略](2026-09-09-consumer-owned-startup-strictness.zh.md))。
2. `dsh-client-modules` 的 node 半(该包是双面的:浏览器半就是模块表)使用 Host face import 时相同的 `name` 与所属 tree `baseUrl` 解析每个 live Loader entry,再读取最近归属 package.json 的 `dsh.client` 声明并组合出 `window.__DSH_BOOT__`:`{ rev, entries: [{ id, url, rev, inject?, immediately?, external? }], batches: [{ phase, url, rev, entries }] }`。即使 overlay 指向相对的 source 或 built entry 文件,manifest 包名仍是浏览器模块身份。若不同的 active Loader source 解析到同一包名,组合会失败;一个来源卸载后,仍存活的来源无需重启 fiber 即可提供该 row。Row 的三个可选字段都来自 manifest,永不人肉抄写。组合会把被请求的动态图 row 排到消费者之前、拒绝同步请求环,并把每个 row 恰好分配给一个初始批次。它会拒绝没有已构建 `./client` bundle 的已声明插件,并把它们的 package/path 行归到一条源码构建要求下;畸形声明字段同样会让激活失败,Host 检查会从 FAILED fiber 报告这两类错误。
3. 扫描是单包增量——不存在全量重扫代码路径。每次 cordis `internal/plugin` 发射把该 fiber 的 entry 名标脏(无 entry 的 fiber O(1) 丢弃);微任务 flush 把每个脏名对账 live loader entries,包元数据(含「非 client 包」的否定结论)按 entry 名与所属 tree base URL 缓存至进程结束,bundle 重哈希只经 `rebuilt(id)` 可达。激活趟从当前 entries 灌同一脏集合并同步 flush,初扫与稳态共享一条实现。初始 row 使用不透明的进程 nonce 加序号,不对其产物求哈希;启动 combo revision 对合并脚本输入及 indexed map 求哈希,row 与批次描述再共同哈希进 `graph.rev`。图类型单源在 modules 包的 `./client` 出口——webserver 对图一无所知;modules 会注册 combo 路由并贡献结构化 index 注入行。
为什么名册是 yml 行而不是扫描?因为哪些插件组合进一次部署是组合决策,不是包属性——一个在仓库中声明了 dsh.client 的包,不代表这次部署要挂载它,扫描发现无从替人做这个决定;node 半只扫描配置树实际挂载了的东西。
**第一阶段——模块面。**注入的 HTML 以 queue 模式安装 `window.__ModuleLoader__`,开始预加载所有 application combo URL,以阻塞式 classic script 依次执行所有 bootstrap combo URL,赋值 `window.__DSH_BOOT__`,然后启动 Vite 主模块。内核把原始图和外壳 seed 传给 facade 的 `create()`。Facade 移除 modules registration,用拒绝全部 external 的 bootstrap `require` 将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造系统、记忆化自身 exports、在模块闭包中保留该实例,并把同一 facade 切换到 live registration。随后内核并行预取每个 `immediately` row;同一 application combo 中的 row 共享一次执行,不同 combo 会在 immediate row、被请求依赖或普通 entry import 首次触及时独立加载。预取失败在这里被吞下,因为第二阶段 import 会重试并拥有那次大声失败。`immediately` 仍是 registration barrier,不是包身份。
**第一阶段——模块面。**注入的 HTML 以 queue 模式安装 `window.__ModuleLoader__`,开始预加载所有 application combo URL,以阻塞式 classic script 依次执行所有 bootstrap combo URL,赋值 `window.__DSH_BOOT__`,然后启动 Vite 主模块。内核把原始图和外壳 seed 传给 facade 的 `create()`。Facade 移除 modules registration,用拒绝全部 external 的 bootstrap `require` 将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造并返回系统、记忆化自身 exports,并把同一 facade 切换到 live registration。随后内核并行预取每个 `immediately` row;同一 application combo 中的 row 共享一次执行,不同 combo 会在 immediate row、被请求依赖或普通 entry import 首次触及时独立加载。预取失败在这里被吞下,因为第二阶段 import 会重试并拥有那次大声失败。`immediately` 仍是 registration barrier,不是包身份。
**第二阶段——插件面。**
1. 内核挂载 vendored Loader,在任何 entry 存在之前就把模块系统注入为 `internal`。顺序有讲究:`tree.import` 的裸 import 兜底分支在浏览器里绝不能跑到。
2. 它统一创建每个 graph row。Import modules row 会返回记忆化的 bootstrap exports,其 `apply()` 把闭包中的系统提供为 `ctx.modules`;需要该 service 的 row 会保持 PENDING 直至此时,因此 modules row 无需特殊创建位置。渲染组装是由 `dsh-client-ui-renderer` 提供的普通 host graph row;内核不追加组装伪 entry。
2. 它统一创建每个 graph row。Import modules row 会返回记忆化的 bootstrap exports,其 `apply()` 读取该树的 `Loader.internal`,把同一个实例提供为 `ctx.modules`;需要该 service 的 row 会保持 PENDING 直至此时,因此 modules row 无需特殊创建位置。渲染组装是由 `dsh-client-ui-renderer` 提供的普通 host graph row;内核不追加组装伪 entry。
3. Graph 顺序治理同步 factory 可用性;Cordis 激活与之独立,仍经服务等待推进。
4. `settled` = 每个 entry 已创建 + `loader.await()` 完全停稳 + 一次全 ACTIVE 扫描。扫描列出每个 import 失败、FAILED 或 PENDING 的 fiber 及其缺失的服务。它存在的理由:cordis 的 inject 等待没有超时——这次扫描就是大声失败的兜底线。
5. 不依赖框架的 loading 页经 `internal/status` 投影真实 fiber 状态。检查完成后,内核调用 `ctx.uiRenderer.mount(container)`,一次切换到真实 UI。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-24-single-harness-home-resolver.md
2026-07-24-single-harness-home-resolver.md: 02578674b4547bae54b2edce99c6c4ef53748588
2026-07-24-single-harness-home-resolver.zh.md: 5764be3f85901196cd2a2db77d1fc3838668163f
2026-07-24-single-harness-home-resolver.md: 1191c52bfa222b807e9aa301b250d2efcc8b0953
2026-07-24-single-harness-home-resolver.zh.md: cefa124b7cb46bab92157af34576f59d3053da9f
@@ -23,6 +23,8 @@ explicit configured path > $DSH_HOME > ~/.dsh
An empty or whitespace-only `$DSH_HOME` is treated as unset; otherwise `resolve('')` would silently place the home at the current working directory. The harness keeps all user data under one root; there is no XDG config/data/cache split. `dshHomePath(...segments)` joins deployment-owned children onto that root, and `dsh-app-boot` exposes it to Loader `!!js` config expressions before mounting entries, so shipped compositions derive `sessions` and `storages` without copying the resolver. `dshHomeDisplay()` names a resolved root symbolically for user-facing paths — `~/.dsh` for the default home, `$DSH_HOME` for any configured home — so the user-global `AGENTS.md` label never leaks an absolute machine path. It replaces agent-instructions's bespoke default-vs-`$DSH_HOME` check.
`dshCachePath(...segments)` derives paths below the resolved home's `cache` directory. An initial `{ dshHome }` option preserves a provider's explicit home override. It resolves paths without creating directories; callers own directory creation. `attachment-local` uses this helper for regenerable request-image variants while retaining durable attachment objects in their versioned storage tree, so clearing the cache cannot remove Session attachments. Existing request-image cache entries are left in place and are not read or copied; a cache miss regenerates the variant from its durable attachment.
`@deepseek-ai/dsh-home` is deleted. Home-owning providers and boot packages import `resolveDshHome` from `dsh-home-paths`; composition bundles contain only the resolved configuration rows.
`dsh-telemetry` and its separate home policy are absent under the [SDK project toolchain removal](../../archived/simplification/2026-08-11-remove-sdk-project-toolchain.md), leaving this resolver as the sole home policy.
@@ -23,6 +23,8 @@ explicit configured path > $DSH_HOME > ~/.dsh
空或仅含空白的 `$DSH_HOME` 被当作未设置处理;否则,`resolve('')` 会悄悄把 home 落在当前工作目录。harness 把所有用户数据都放在同一个根目录下;不存在 XDG 的 config/data/cache 拆分。`dshHomePath(...segments)` 将部署负责的子路径拼接到该根目录下,`dsh-app-boot` 在挂载条目前向 Loader `!!js` 配置表达式暴露它,因此出厂组合无需复制解析器即可派生 `sessions` 和 `storages`。`dshHomeDisplay()` 为面向用户的路径以符号形式命名已解析的根目录——默认 home 显示为 `~/.dsh`,任何已配置的 home 显示为 `$DSH_HOME`——这样用户全局的 `AGENTS.md` 标签就绝不会泄露机器上的绝对路径。它取代了 agent-instructions 中自定义的「默认值 vs `$DSH_HOME`」判断。
`dshCachePath(...segments)` 在解析出的主目录下的 `cache` 目录中派生路径。首个 `{ dshHome }` 选项保留提供方显式配置的主目录覆盖值。它只解析路径,不创建目录;目录创建由调用方负责。`attachment-local` 将此函数用于可重新生成的请求图片版本,持久附件对象仍保留在其带版本的存储树中,因此清空缓存不会删除 Session 附件。已有请求图片缓存条目保留在原处,不再读取或复制;缓存未命中时从持久附件重新生成请求版本。
`@deepseek-ai/dsh-home` 被删除。拥有 home 配置的提供方与 boot 包从 `dsh-home-paths` 导入 `resolveDshHome`;组合包只包含解析后的配置行。
`dsh-telemetry` 及其独立 home 策略已随 [SDK 项目工具链移除](../../archived/simplification/2026-08-11-remove-sdk-project-toolchain.md)一并消失,因此该解析器是唯一的 home 策略。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-24-web-config-tree-boot-and-transport-layering.md
2026-07-24-web-config-tree-boot-and-transport-layering.md: e7c0781e1504bce12a8b0d2197bd47b37d3e873a
2026-07-24-web-config-tree-boot-and-transport-layering.zh.md: 9a0affa397151a3994c5731bebda4b45a4139b6c
2026-07-24-web-config-tree-boot-and-transport-layering.md: c8a1db3cd7bd715d8912b22a0b41a9ef660c1316
2026-07-24-web-config-tree-boot-and-transport-layering.zh.md: 41fcf65b06829199c203faa806bd17a310b68bcf
@@ -12,7 +12,7 @@ English | [中文](2026-07-24-web-config-tree-boot-and-transport-layering.zh.md)
## Decision
**Composition is one flat assembled tree.** `apps/cli/config/base.cordis.yml` plus `apps/cli/config/web.cordis.yml` holds every row — the host runtime (32 rows), the `api-gateway` row, the `webserver` row, and the `dsh.client` rows (the browser roster; the modules row is simultaneously a host row). No spine bundle: every plugin is one row and every config field is yml-editable. That stance later became repository-wide, with the rows both surfaces share factored into `apps/cli/config/base.cordis.yml` and each surface reduced to an overlay ([shared-base overlays](../../archived/simplification/2026-07-29-shared-base-config-overlays.md)). The `dsh-client-hmr` row is an ordinary always-on bundle row (originally appended in code by `--dev`; the flag is retired). Row order carries no load semantics; activation is service-availability driven. The shared audit rejects imports with no fiber, awaits only failed fibers to recover original activation errors, and reports services that leave a fiber `PENDING`; before throwing, it marks those exact rejection reasons through one process checkpoint so `installFailLoud` coalesces Loader's duplicate notification while unrelated unhandled rejections remain fatal. The Node app-boot artifact embeds `@cordisjs/plugin-include` while leaving `@cordisjs/plugin-loader` external, so the include's `EntryTree` and the host bind to one Loader peer instead of splitting a config tree across two Loader implementations.
**Composition is one flat assembled tree.** `apps/cli/config/base.cordis.yml` plus `apps/cli/config/web.cordis.yml` holds every row — the host runtime (32 rows), the `api-gateway` row, the `webserver` row, and the `dsh.client` rows (the browser roster; the modules row is simultaneously a host row). No spine bundle: every plugin is one row and every config field is yml-editable. That stance later became repository-wide, with the rows both surfaces share factored into `apps/cli/config/base.cordis.yml` and each surface reduced to an overlay ([shared-base overlays](../../archived/simplification/2026-07-29-shared-base-config-overlays.md)). The `dsh-client-hmr` row is an ordinary always-on bundle row (originally appended in code by `--dev`; the flag is retired). Row order carries no load semantics; activation is service-availability driven. `auditStartupEntries` reports failed imports, reads failed fibers for original activation errors, and lists services that leave a fiber `PENDING`. The [startup policy](2026-09-09-consumer-owned-startup-strictness.md) makes optional failures warnings and required failures fatal. Reported rejection reasons stay marked through one process checkpoint so `installFailLoud` coalesces Loader's duplicate notification while unrelated unhandled rejections remain fatal. The Node app-boot artifact embeds `@cordisjs/plugin-include` while leaving `@cordisjs/plugin-loader` external, so the include's `EntryTree` and the host bind to one Loader peer instead of splitting a config tree across two Loader implementations.
**Boot glue is a class pair.** `AppCLIEntry` (apps/cli) and `AppWebEntry` (the shell kernel) hold only what must exist independently of cordis: argv facts, the composed patch set, the parsed boot manifest, the module system instance, loading-page handles — everything else lives in plugins. `AppCLIEntry.run()` is three stages: layered env (ambient > cwd `.env` > `$DSH_HOME/.env`, closing the defect above) → patch composition → Loader include boot plus the activation audit. `AppWebEntry.run()` mirrors it browser-side: parse `window.__DSH_BOOT__` into a `BootManifest` (two views: npm-package rows for the module table, cordis-plugin rows for entry composition; malformed wire throws), build the module system, render the loading page, prefetch the `immediately` tier in parallel with Context/Loader setup, **await the prefetch before creating entries** (materialization is `tree.import`'s synchronous require, unprotected by fiber inject waiting; cross-package require edges such as i18n → runtime/client need every immediately-tier factory registered first — an empirically found 10–25% boot race otherwise), adopt the modules entry, create the graph rows, settle, sweep.
@@ -12,7 +12,7 @@ Status: implemented
## 决策
**组合结果是一棵平铺配置树。** `apps/cli/config/base.cordis.yml` 与 `apps/cli/config/web.cordis.yml` 共同持有全部行——host 运行时(32 行)、`api-gateway` 行、`webserver` 行、`dsh.client` 行(浏览器 roster;modules 行同时是 host 行)。不做主干 bundle:每插件一行、每个 config 字段 yml 可改。这一立场后来推广到全仓:两个 surface 共享的配置项被抽取进 `apps/cli/config/base.cordis.yml`,各 surface 则收敛为一份 overlay([共享 base overlay](../../archived/simplification/2026-07-29-shared-base-config-overlays.md))。`dsh-client-hmr` 行是普通的始终启用的 bundle 行(最初由 `--dev` 在代码中追加;该旗标已废除)。行序无装载语义;激活由服务可用性驱动。共享 audit 会拒绝没有 fiber 的 import、仅等待失败的 fiber 以恢复原始激活错误,并报告让 fiber 停在 `PENDING` 的服务;抛出错误前,审计会通过一个进程级检查点标记这些 rejection 的确切原因,从而让 `installFailLoud` 将 Loader 的重复通知合并为一次,而无关的未处理 rejection 仍然致命。Node app-boot 产物内嵌 `@cordisjs/plugin-include`,但将 `@cordisjs/plugin-loader` 保持为外部依赖,因此 include 的 `EntryTree` 与 host 会绑定到同一个 Loader peer,而不会让一棵配置树横跨两个 Loader 实现。
**组合结果是一棵平铺配置树。** `apps/cli/config/base.cordis.yml` 与 `apps/cli/config/web.cordis.yml` 共同持有全部行——host 运行时(32 行)、`api-gateway` 行、`webserver` 行、`dsh.client` 行(浏览器 roster;modules 行同时是 host 行)。不做主干 bundle:每插件一行、每个 config 字段 yml 可改。这一立场后来推广到全仓:两个 surface 共享的配置项被抽取进 `apps/cli/config/base.cordis.yml`,各 surface 则收敛为一份 overlay([共享 base overlay](../../archived/simplification/2026-07-29-shared-base-config-overlays.md))。`dsh-client-hmr` 行是普通的始终启用的 bundle 行(最初由 `--dev` 在代码中追加;该旗标已废除)。行序无装载语义;激活由服务可用性驱动。`auditStartupEntries` 报告 import 失败、读取失败 fiber 的原始激活错误,并列出让 fiber 停在 `PENDING` 的服务。[启动策略](2026-09-09-consumer-owned-startup-strictness.zh.md)将 optional failure 作为 warning,将 required failure 视为致命错误。已报告的 rejection 原因会保持标记至一个进程级检查点,使 `installFailLoud` 合并 Loader 的重复通知,而无关的未处理 rejection 仍然致命。Node app-boot 产物内嵌 `@cordisjs/plugin-include`,但将 `@cordisjs/plugin-loader` 保持为外部依赖,因此 include 的 `EntryTree` 与 host 会绑定到同一个 Loader peer,而不会让一棵配置树横跨两个 Loader 实现。
**boot 胶水由两个类组成。** `AppCLIEntry`(apps/cli)与 `AppWebEntry`(壳内核)只持有那些必须独立于 cordis、提前存在的东西:argv 事实、合成的 patch 集、解析出的 boot manifest(元数据清单)、模块系统实例、loading 页句柄——其余一律进插件。`AppCLIEntry.run()` 三段:分层 env(ambient > cwd `.env` > `$DSH_HOME/.env`,顺手关掉上述缺陷)→ patch 合成 → Loader include boot 加 activation audit。`AppWebEntry.run()` 在浏览器侧镜像它:把 `window.__DSH_BOOT__` 解析成 `BootManifest`(双视角:npm 包行给模块表、cordis 插件行给 entry 组合;畸形 wire 大声抛)、建模块系统、渲染 loading 页、immediately 层预取与 Context/Loader 准备并行、**create entry 之前等预取齐**(物化是 `tree.import` 的同步 require,不受 fiber inject 等待保护;i18n → runtime/client 这类跨包 require 边要求 immediately 层工厂全部注册完——否则有实测 10–25% 的 boot 竞态)、收编 modules entry、逐一创建图行、settle、sweep。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-28-portable-execution-world-consumers.md
2026-07-28-portable-execution-world-consumers.md: 787fe341e58cc212c99e0f35f07eea8e83daf000
2026-07-28-portable-execution-world-consumers.zh.md: a558a5af64437b8743e741ace4ccf27079501721
2026-07-28-portable-execution-world-consumers.md: 5a687a552f322f3929b3f3937caf36c3373c05e7
2026-07-28-portable-execution-world-consumers.zh.md: 6bc0e20bd29d0b0a01068c004655cc90ac3b6699
@@ -16,6 +16,8 @@ Ordinary pipes do not cover one requirement. A persistent terminal needs PTY all
`ctx.fs` and `ctx.subprocess` together define one execution world. Providers mounted together must describe the same path namespace, executables, processes, and terminal sessions; higher capabilities consume those two interfaces rather than name the provider.
Foreground Bash and PowerShell calls use one executor deadline for asynchronous confinement preparation and process execution. Preparation expiry returns an outcome without process-exit or signal facts, and late argv cannot start a process; background preparation follows caller cancellation only. The protected execution result tells subclasses whether the subprocess provider was called, so enforcement facts are attached only after publication.
The filesystem interface owns the path facts that another capability needs without exposing its opaque target identity: a canonical process path, canonical `file:` URI, and containment. Existing whole and streaming text operations remain filesystem-owned; protocol consumers enforce their own retention limits while consuming the stream.
The subprocess interface owns executable lookup and process primitives: ordinary raw or collected process spawning and `spawnTerminal()`. An ordinary handle keeps target identity private: `.done` reports the direct target, while `terminate()` and `waitForExit()` control and observe the same provider-managed range. The [native-containment decision](2026-08-28-subprocess-native-containment.md) owns local Linux scopes, Windows Jobs, and their disclosed fallbacks. The terminal operation is one deep primitive whose handle owns text I/O, foreground groups, signalling, and one awaited TERM-to-KILL operation that settles in-flight handle calls and reaches quiescence for every member of its provider-owned range; an observational fallback limits that range to identities it can still observe. Its signal cancels allocation only; the published handle owns its lifetime. Prompt detection, idle inference, scrollback, sandbox policy, and owner lifecycle remain in the PTY consumer.
@@ -26,19 +28,15 @@ Generic consumers use that execution world:
- `dsh-lsp-stdio` reads and contains source through `ctx.fs`, resolves and launches language servers through `ctx.subprocess`, and carries provider-owned file URIs through initialization and result rendering. One provider-lifetime signal aborts filesystem and protocol work during disposal, including workspace lookup before queue ownership; its JSON-RPC, pooling, synchronization, and normalization stay unchanged.
- `dsh-terminal-bash` maps persistent-shell semantics onto `ctx.subprocess.spawnTerminal()`. The local `node-pty` and process-inspection implementation moves into `dsh-subprocess-local`; another subprocess provider supplies the same primitive. `danger-full-access` needs no `ctx.sandbox`; a confined mode requires a same-world sandbox provider and fails before spawn when none is mounted. Prompt and silence evidence collected during asynchronous pre-write inspection is discarded when the provider write begins. Cancellation retains the send reservation while an in-flight write settles and then signals the foreground group, so late bytes or the signal cannot target a successor; an in-flight readiness poll cannot release that reservation, and a rejected write sends no signal. The absolute deadline remains armed throughout cancellation. A signal failure becomes terminal transport failure. Completion of a stale inspection resumes polling for the current send. Startup cancellation begins terminal rollback without waiting for a stalled readiness or signalling call. Close rejects new public signals and delegates provider-managed session quiescence to the handle's awaited termination operation.
## E2B POC boundary
## Remote provider ownership
The opt-in E2B realization has exactly three provider-specific packages under `packages/e2b/`: `dsh-e2b` creates one sandbox and deletes it on timeout or disposal, `dsh-fs-e2b` implements `ctx.fs`, and `dsh-subprocess-e2b` implements `ctx.subprocess` over E2B Commands, PTYs, and remote Linux process groups. The two adapters obtain the sole SDK handle from the owner and never create private sandboxes.
The [E2B provider removal](../simplification/2026-09-11-remove-e2b-providers.md) supersedes the E2B realization of this decision. The filesystem/subprocess agreement and asynchronous terminal contracts remain in force for remote implementations. The [POSIX SSH providers](2026-09-11-posix-ssh-runtime.md) realize them through one installed helper and independent stream channels.
E2B owns the mutable filesystem, managed command and Bash processes, terminal allocation and terminal-session groups, language-server processes and source reads, and adapter-private files under `.dsh-e2b`. The host owns Cordis and plugin objects, the agent loop, agent/session/goal state, session logs and persistence, LLM calls, prompts and tools, authority, skills, subagent orchestration, PTY buffers and readiness, LSP protocol state, and E2B SDK/network buffers. The overlay neither uploads nor synchronizes the host workspace.
The adapters retain only substrate mechanics. Filesystem canonicalization crosses the SDK's decoded command transport as strict base64-encoded NUL framing; streamed reads leave byte ceilings with consumers. Subprocess command output and environment snapshots use ASCII/base64 where SDK chunk decoding would otherwise lose bytes, while private control shells isolate profiles and later launches blank discovered credential-shaped names. Process and terminal cleanup uses remote groups and proves quiescence before settlement.
Sandbox state is deliberately ephemeral: timeout and disposal delete the remote files and unmanaged state. The POC adds no reconnect or pause/leave retention, session-persistence backend, template builder, volume, snapshot, network-policy layer, sandbox catalog, workspace synchronization, durable remote handles, or whole-harness execution.
A remote provider owns mutable files, command and terminal processes, language-server processes, and provider-private runtime files. The host owns Cordis and plugin objects, the agent loop, agent/session/goal state, session logs and persistence, model transport, authority, skills, subagent orchestration, terminal readiness and LSP protocol state. Moving execution does not imply workspace synchronization or durable remote handles.
## Verification
Focused package suites pin sandbox lifecycle, canonical path framing, filesystem metadata and atomic versions, subprocess publication/rollback, terminal text I/O and session cleanup, output limits, cancellation, disposal, and invariant registration. A credential-gated Loader composition exercises the same three-package provider through source imports and built exports, including FS/Bash visibility, hostile login profiles, byte-split UTF-8 output, process and terminal cleanup, LSP queries, host-workspace isolation, and final sandbox deletion.
The local filesystem, subprocess, terminal and LSP suites cover path identity, executable lookup, managed cleanup, output limits, cancellation and disposal. Terminal consumer tests retain delayed provider writes, foreground inspection and signalling to verify asynchronous ownership without requiring a remote service.
## Alternatives considered
@@ -60,14 +58,14 @@ Focused package suites pin sandbox lifecycle, canonical path framing, filesystem
**Implement remote filesystem operations only through shell commands.** Rejected because that discards structured filesystem identity, errors, streaming, version guards, and atomic mutation semantics already consumed by the file tools.
**Add a generic distributed-runtime abstraction or reconnect live handles.** Rejected because the existing capability seams carry the demonstrated contracts, while remote identity alone cannot reconstruct callbacks, pending promises, authority, protocol state, or output cursors. A new layer would speculate about persistence and synchronization beyond the POC.
**Add a generic distributed-runtime abstraction or reconnect live handles.** Rejected because the existing capability seams carry the demonstrated contracts, while remote identity alone cannot reconstruct callbacks, pending promises, authority, protocol state, or output cursors. A new layer would speculate about persistence and synchronization beyond the demonstrated consumer contracts.
## Consequences
A remote execution provider implements only its shared sandbox owner plus filesystem and subprocess adapters. Bash, PTY, and LSP compose above them, so fixes to those capabilities remain provider-neutral.
A remote execution family supplies its shared connection owner and matching filesystem, subprocess and, for confined calls, sandbox providers. Bash, PTY, and LSP compose above them, so fixes to those capabilities remain provider-neutral.
The fundamental interfaces are wider, and a filesystem/subprocess pair must agree on one execution world. The added operations are limited to facts and lifecycle mechanics that current generic consumers require; model schemas, protocol framing, readiness policy, and presentation do not leak into the providers.
The local implementation absorbs `node-pty` and platform process inspection because it owns local terminal mechanics. On supported Linux hosts, the user-systemd scope retains descendants that call `setsid` or reparent, while process inspection continues to own foreground attribution and synchronous fallback evidence. Other hosts use the observational teardown: disposal sweeps descendants before and after terminating the top-level shell, waits for exact PID-identity-fenced descendants retained during foreground inspection, and retains Linux session members that survive top-level exit. macOS cannot enumerate a POSIX session after its leader exits, so a child that reparents between inspection snapshots remains an explicit local-provider limitation rather than a reason to move process mechanics back into the PTY consumer.
The E2B composition demonstrates that a shared sandbox owner plus filesystem and subprocess adapters are sufficient to move the mutable coding world off-host while leaving higher capabilities provider-neutral. Its POC limits remain explicit: the SDK retains complete command transport in host memory, remote startup cannot publish a PID synchronously, exact terminal stdin-wait and independent signal facts are unavailable, numeric PID/PGID operations are not identity-fenced, the initial environment probe cannot hide unknown sandbox-default secrets from already-running same-UID processes, and adapter artifacts remain until sandbox deletion. These are provider constraints, not justification for compatibility shims or more E2B packages.
Remote implementations must preserve the existing consumer contracts or reject unsupported operations explicitly. Transport-specific buffering, identity and disconnect limits belong to the provider; they do not justify duplicate Bash, PTY or LSP implementations.
@@ -16,6 +16,8 @@ Status: implemented
`ctx.fs` 与 `ctx.subprocess` 共同定义一个执行世界。共同挂载的提供方必须描述相同的路径命名空间、可执行文件、进程和终端会话;上层能力消费这两个接口,而不引用具体提供方。
前台 Bash 和 PowerShell 调用由同一执行器 deadline 覆盖异步 confinement 准备与进程执行。准备超时返回没有进程退出或信号事实的结果,晚到的 argv 不能启动进程;后台准备仍只跟随调用方取消。子类通过受保护的执行结果区分是否已调用 subprocess provider,只有发布后才附加 enforcement 事实。
文件系统接口负责其他能力需要的路径事实,同时不公开其不透明目标身份:规范化进程路径、规范化 `file:` URI 和包含关系。现有完整文本与流式文本操作仍归文件系统负责;协议消费方在消费流时执行各自的保留上限。
进程管理接口负责可执行文件查找与进程原语:以原始或收集模式 spawn 普通进程,以及 `spawnTerminal()`。普通句柄把 target identity 保持为私有事实:`.done` 报告 direct target,`terminate()` 与 `waitForExit()` 则控制并观察同一个由提供方管理的范围。[原生 containment 决策](2026-08-28-subprocess-native-containment.zh.md)负责本地 Linux scope、Windows Job 及其已声明的 fallback。终端操作是一项深层原语,其句柄负责文本 I/O、前台进程组、信号发送,以及一项须等待的 TERM→KILL 操作;该操作会结算所有在途句柄调用,并使提供方拥有的范围中每个成员完全停稳;观察型 fallback 只能把该范围限制为它仍可观察到的 identity。其信号只取消分配;句柄一经发布,便负责自身生命周期。提示符检测、空闲推断、scrollback、沙箱策略和所有者生命周期仍由 PTY 消费方负责。
@@ -26,19 +28,15 @@ Status: implemented
- `dsh-lsp-stdio` 通过 `ctx.fs` 读取源文件并验证包含关系,通过 `ctx.subprocess` 解析和启动语言服务器,并让由提供方负责的文件 URI 贯穿初始化与结果渲染。一个提供方生命周期信号会在资源释放期间中止文件系统与协议操作,包括取得队列所有权之前的工作区查找;其 JSON-RPC、池化、同步和规范化保持不变。
- `dsh-terminal-bash` 把持久 shell 语义映射到 `ctx.subprocess.spawnTerminal()`。本地 `node-pty` 与进程检查实现移入 `dsh-subprocess-local`;其他进程管理提供方则提供相同原语。`danger-full-access` 不需要 `ctx.sandbox`;受限模式要求同一执行世界中存在沙箱提供方,未挂载时会在 spawn 前失败。提供方开始写入时,系统会丢弃异步写入前检查期间收集的提示符与静默证据。取消会在在途写入结算期间保留发送预留,随后向前台进程组发送信号,因此延迟字节和该信号都无法落到后续发送;在途就绪检查无法释放该预留,写入被拒绝时也不会发送信号。绝对截止时间会在整个取消期间保持启用。信号发送失败会成为终结性传输失败。陈旧检查完成后,会针对当前发送恢复轮询。启动取消会立即开始终端回滚,而不等待停滞的就绪检查或信号发送调用。关闭操作会拒绝新的公开信号,并把由提供方管理的会话完全停稳委托给句柄上须等待的终止操作。
## E2B POC 边界
## 远程提供方的职责
可选启用的 E2B 实现在 `packages/e2b/` 下恰好只有三个提供方专用包:`dsh-e2b` 创建一个沙箱,并在超时或资源释放时将其删除;`dsh-fs-e2b` 实现 `ctx.fs`;`dsh-subprocess-e2b` 基于 E2B Commands、PTY 和远程 Linux 进程组实现 `ctx.subprocess`。两个适配器都从所有者取得唯一的 SDK 句柄,绝不创建私有沙箱。
[E2B 提供方移除决策](../simplification/2026-09-11-remove-e2b-providers.zh.md)取代本决策中的 E2B 实现。文件系统/子进程约定与异步终端约定继续适用于远程实现。[POSIX SSH 提供方](2026-09-11-posix-ssh-runtime.zh.md)通过一个已安装辅助程序与独立流通道实现这些约定。
E2B 负责可变文件系统、受管命令与 Bash 进程、终端分配与终端会话组、语言服务器进程与源文件读取,以及 `.dsh-e2b` 下的适配器私有文件。宿主负责 Cordis 与插件对象、agent loop(智能体循环)、agent 状态、会话状态与目标状态、会话日志与持久化、LLM(大语言模型)调用、提示词与工具、权限、skill(技能)、subagent 编排、PTY 缓冲区与就绪状态、LSP 协议状态,以及 E2B SDK/网络缓冲区。该叠加层既不上传,也不同步宿主工作区。
适配器只保留执行基底机制。文件系统规范化以严格的 base64 加 NUL 分帧穿过 SDK 已解码的命令传输;流式读取把字节上限留给消费方执行。进程管理命令输出与环境快照采用 ASCII/base64,避免 SDK 分片解码丢失字节;私有控制 shell 隔离 profile,后续启动会把已发现且名称呈凭据特征的环境变量置空。进程与终端清理使用远程进程组,并在结算前证明完全停稳。
沙箱状态有意保持短暂:超时与资源释放会删除远程文件和非托管状态。该 POC 不提供重新连接、pause/leave 保留、会话持久化后端、模板构建器、卷、快照、网络策略层、沙箱目录、工作区同步、持久远程句柄,也不会在其中运行整个 harness。
远程提供方负责可变文件、命令与终端进程、语言服务器进程,以及提供方私有运行时文件。宿主负责 Cordis 与插件对象、智能体循环、智能体/会话/目标状态、会话日志与持久化、模型传输、权限、技能、子智能体编排、终端就绪判断和 LSP 协议状态。移动执行位置不意味着工作区同步或持久远程句柄。
## 验证
聚焦的包测试套件锁定了沙箱生命周期、规范化路径分帧、文件系统元数据与原子版本、进程管理发布/回滚、终端文本 I/O 与会话清理、输出上限、取消、资源释放和不变式注册。一项受凭据门控的 Loader 组合通过源代码导入与构建后导出运行同一套三包提供方组合,其中包括 FS/Bash 可见性、恶意登录 profile、跨字节边界拆分的 UTF-8 输出、进程与终端清理、LSP 查询、宿主工作区隔离,以及最终沙箱删除。
本地文件系统、子进程、终端与 LSP 测试覆盖路径身份、可执行文件查找、受管清理、输出上限、取消与资源释放。终端消费方测试保留延迟的提供方写入、前台检查和信号发送,以便在无需远程服务时验证异步所有权。
## 考虑过的替代方案
@@ -60,14 +58,14 @@ E2B 负责可变文件系统、受管命令与 Bash 进程、终端分配与终
**只通过 shell 命令实现远程文件系统操作。** 不予采纳,因为这会丢弃现有文件工具已消费的结构化文件系统身份、错误、流式输出、版本保护和原子变更语义。
**新增通用分布式运行时抽象,或重新连接活跃句柄。** 不予采纳,因为现有能力 seam 已承载经证实的约定,而仅凭远程身份无法重建回调、待处理 promise、权限、协议状态或输出游标。新增一层只会推测 POC 边界之外的持久化与同步问题。
**新增通用分布式运行时抽象,或重新连接活跃句柄。** 不予采纳,因为现有能力 seam 已承载经证实的约定,而仅凭远程身份无法重建回调、待处理 promise、权限、协议状态或输出游标。新增一层只会推测 已验证消费方约定之外的持久化与同步问题。
## 后果
远程执行提供方只需实现共享沙箱所有者,以及文件系统与进程管理适配器。Bash、PTY 与 LSP 组合在这些适配器之上,因此对这些能力的修复仍与提供方无关。
远端执行家族提供共享连接管理器及配套文件系统、子进程提供方;受限调用还需要沙箱提供方。Bash、PTY 与 LSP 组合在这些适配器之上,因此对这些能力的修复仍与提供方无关。
基础接口更宽,一对文件系统/进程管理提供方必须在同一个执行世界上保持一致。新增操作仅限当前通用消费方所需的事实与生命周期机制;模型 schema、协议分帧、就绪策略和呈现不会渗入提供方。
本地实现承接 `node-pty` 和平台进程检查,因为它负责本地终端机制。在受支持的 Linux 宿主上,user-systemd scope 会保留调用 `setsid` 或发生 reparent 的后代,进程检查则继续负责前台归属与同步 fallback 证据。其他宿主使用观察型拆卸:dispose(资源释放)会在终止顶层 shell 前后清理后代进程,等待前台检查期间保留下来且受精确 PID 身份围栏保护的后代进程,并继续追踪在顶层进程退出后仍存活的 Linux 会话成员。macOS 无法在 POSIX 会话 leader 退出后枚举该会话,因此在两次检查快照之间重新设定父进程的子进程仍是明确的本地提供方限制,而不是把进程机制移回 PTY 消费方的理由。
E2B 组合证明,共享沙箱所有者加上文件系统与进程管理适配器,就足以在保持上层能力与提供方无关的同时,把可变编码世界移出宿主。其 POC 限制仍明确在案:SDK 会把完整命令传输内容保留在宿主内存中;远程启动无法同步发布 PID;无法获得精确的终端 stdin 等待状态与独立信号事实;基于数值 PID/PGID 的操作没有身份围栏;初始环境探测无法向已在运行的同 UID 进程隐藏未知的沙箱默认 secret;适配器产物会一直保留到沙箱删除。这些是提供方限制,不是引入兼容性 shim 或更多 E2B 包的理由。
远程实现必须保留现有消费方约定,或明确拒绝不支持的操作。传输特定的缓冲、身份与断线限制归提供方负责;这些限制不是复制 Bash、PTY 或 LSP 实现的理由。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-29-projected-token-usage-and-request-context.md
2026-07-29-projected-token-usage-and-request-context.md: f96257243bef91ff6a73418231de5e777d8edb2e
2026-07-29-projected-token-usage-and-request-context.zh.md: 7365d5d816f9f9b324f3e3d3b4db0d3346851bdf
2026-07-29-projected-token-usage-and-request-context.md: 10056ac8ef85562035d4fc4591e90aec603a7dd8
2026-07-29-projected-token-usage-and-request-context.zh.md: 0e4cd68791f0ee65dd700b64493be3c33b78a920
@@ -58,4 +58,4 @@ Token totals stay stable across pagination, compaction, replay, restart, and rec
Occupancy is approximate in the ways documented above. It is available immediately after restore or reconnect, since both fields are durable, at the cost of describing the last recorded request rather than an exact current boundary.
Each session log gains one small `request/context` record per route or advertised-capacity change. Token-meter is the canonical owner of durable usage semantics, including retry-attempt separation in the cumulative projection and the reusable exact attempt/Turn fold; Web Chat only selects a complete loaded Turn and renders the fold result. The TUI retains its live per-step map because it does not mount the generic projection seam, and the standalone browser fixture mirrors the unit. Connection and API Gateway carry no token-specific code, own no per-session metrics cache, and perform no measurement. The browser keeps two generic projection values and no connection-local telemetry; streaming text deltas do not force the stats line to recompute or churn layout-observer subscriptions.
Each session log gains one small `request/context` record per route or advertised-capacity change. Token-meter is the canonical owner of durable usage semantics, including retry-attempt separation in the cumulative projection and the reusable exact attempt/Turn fold; Web Chat only selects a complete loaded Turn and renders the fold result. The TUI retains its live per-step map because it does not mount the generic projection seam, and the assembled RemoteMock scenario supplies the same projection values. Connection and API Gateway carry no token-specific code, own no per-session metrics cache, and perform no measurement. The browser keeps two generic projection values and no connection-local telemetry; streaming text deltas do not force the stats line to recompute or churn layout-observer subscriptions.
@@ -58,4 +58,4 @@ token 总量在分页、压缩、回放、重启和重连期间保持稳定,
占用率在上文记录的意义上是近似值。由于两个字段都是持久的,它在恢复或重连后立即可用;代价是它描述的是最后一条已记录的请求,而不是精确的当前边界。
每个会话日志会为每次路由或已公布容量变化增加一条小型 `request/context` 记录。token-meter 是持久用量语义的正典所有方,包括累计投影中的重试 attempt 分离,以及可复用的精确 attempt/Turn fold;Web Chat 只选择已完整加载的 Turn 并渲染 fold 结果。TUI 未挂载通用投影 seam,因此保留自己的实时逐步骤 map,而独立浏览器 fixture(测试前置数据)会镜像该单元。Connection 与 API Gateway 不携带任何 token 专用代码,不拥有逐会话指标缓存,也不执行测量。浏览器只保留两个通用投影值,不保留连接本地的遥测数据;流式文本增量不会迫使统计行重新计算或反复替换布局 observer 订阅。
每个会话日志会为每次路由或已公布容量变化增加一条小型 `request/context` 记录。token-meter 是持久用量语义的正典所有方,包括累计投影中的重试 attempt 分离,以及可复用的精确 attempt/Turn fold;Web Chat 只选择已完整加载的 Turn 并渲染 fold 结果。TUI 未挂载通用投影 seam,因此保留自己的实时逐步骤 map,而组装 RemoteMock 场景提供相同投影值。Connection 与 API Gateway 不携带任何 token 专用代码,不拥有逐会话指标缓存,也不执行测量。浏览器只保留两个通用投影值,不保留连接本地的遥测数据;流式文本增量不会迫使统计行重新计算或反复替换布局 observer 订阅。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-30-session-end-seed-log-boundary.md
2026-07-30-session-end-seed-log-boundary.md: aeec2a36d0b1e498591ef509e2e9164f586ed60c
2026-07-30-session-end-seed-log-boundary.zh.md: ceba46a474c402230dbf215a2d53a41d3c027fc2
2026-07-30-session-end-seed-log-boundary.md: 9cf7c7f20701688208e8b1c343904f2a9f2cab9e
2026-07-30-session-end-seed-log-boundary.zh.md: fdf4ec9de610f17e792de98334dd5d37f5e4c355
@@ -8,7 +8,7 @@ English | [中文](2026-07-30-session-end-seed-log-boundary.zh.md)
A plugin that owns a standalone open/close bracket in the session log cannot tell a dead marker from a live one. `compaction/start` … `compaction/end` is the shipped case: on picking up a log whose last compaction event is an unmatched `compaction/start`, "the previous writer died mid-compaction" and "a compaction is running right now" are byte-identical stored history. The owner must either refuse to compact a log that is actually free (wedging the session) or proceed over one that is genuinely busy.
Nothing in the log marked where inherited history ended. `session/created`, `session/disposed`, and `session/flush` are cordis runtime signals, not log events; `agent/session-start` is emit-only. `Session.firstLiveSeq` already held the answer exactly — the seq of this lifecycle's first own write — but only in memory, so a consumer reading stored bytes could not see it.
Nothing in the log marked where inherited history ended. `session/created`, `session/disposed`, and `session/flush` are cordis runtime signals, not log events; `agent/created` is also a non-durable runtime event. `Session.firstLiveSeq` already held the answer exactly — the seq of this lifecycle's first own write — but only in memory, so a consumer reading stored bytes could not see it.
Crash repair does not close the gap and must not: `interruptedTurnClosers` synthesizes turn, step, and tool boundaries because core owns that vocabulary, and `compaction/*` belongs to the compaction seam. A core repair pass that closed plugin brackets would put every plugin's bracket semantics in core.
@@ -50,6 +50,6 @@ Bought: one boundary, written in one place, correct for all six seeded-start pat
Cost: a seeded session's log is one event longer, including an empty resumed log. Seq expectations move with that boundary. Two updates are load-bearing rather than mechanical: telemetry's adoption tests assert that capture begins with the current lifecycle's newly appended boundary and excludes the constructor seed, and the property suite's replay invariant is "seed reproduced verbatim, plus one log-only boundary" with idempotence as its own property.
`session/end-seed` joins the on-disk vocabulary. Current v1 requires the validated marker semantics owned by Session; the frozen v0 codec and migration edge own which historical v0 seed layouts remain admissible. The exact inherited cut stays separate from the logical header and is available after a body read.
`session/end-seed` joins the on-disk vocabulary. The current format requires the validated marker semantics owned by Session; the frozen v0 codec and migration edge own which historical v0 seed layouts remain admissible. The exact inherited cut stays separate from the logical header and is available after a body read.
The [queued manual compaction decision](../feature/2026-07-30-queued-manual-compaction.md) now supplies the first consumer. Its tail scan independently finds the unmatched `compaction/start` and newest end-seed, treats only a start after that boundary as live, and clears the invariant trace on the same replay transition. The predicate remains in the compaction package rather than becoming a generic core helper.
@@ -8,7 +8,7 @@ Status: implemented
在会话日志中拥有独立开/闭括号的插件无法区分一个已死的标记和一个存活的标记。`compaction/start` … `compaction/end` 就是已发布的实例:当接手一份日志、而它最后的压缩(compaction)事件是一个未配对的 `compaction/start` 时,「上一个写入方在压缩中途死掉了」与「此刻正有一次压缩在运行」在存储历史中是逐字节相同的。该括号所有方只能二选一:拒绝压缩一份其实空闲的日志(把会话卡死),或者在一份确实繁忙的日志上继续压缩。
日志中没有任何东西标出继承历史在哪里结束。`session/created`、`session/disposed` 与 `session/flush` 是 Cordis 运行时信号,不是日志事件;`agent/session-start` 只发射不落盘。`Session.firstLiveSeq` 本来就精确地持有这个答案——本生命周期第一次自有写入的 seq——但只存在于内存中,因此读取存储字节的消费方看不到它。
日志中没有任何东西标出继承历史在哪里结束。`session/created`、`session/disposed` 与 `session/flush` 是 Cordis 运行时信号,不是日志事件;`agent/created` 同样是不落盘的运行时事件。`Session.firstLiveSeq` 本来就精确地持有这个答案——本生命周期第一次自有写入的 seq——但只存在于内存中,因此读取存储字节的消费方看不到它。
崩溃修复既没有填上这个缺口,也不应该去填:`interruptedTurnClosers` 合成轮次、步骤与工具边界,是因为核心拥有那套词汇表,而 `compaction/*` 属于压缩 seam。一个会关闭插件括号的核心修复流程,等于把每个插件的括号语义都搬进核心。
@@ -50,6 +50,6 @@ Status: implemented
代价:带种子会话的日志长了一个事件,空日志恢复也包括在内。seq 期望会随这条边界移动。两处更新是承重的而非机械的:telemetry 的接管测试断言捕获从当前生命周期新追加的边界开始,并排除 constructor seed;属性测试套件的回放不变式则是「种子逐字节复现,外加一个仅日志边界」,并把幂等性作为独立属性。
`session/end-seed` 加入了落盘词汇表。当前 v1 要求由 Session 拥有的已校验 marker 语义;冻结的 v0 codec 与迁移边负责哪些历史 v0 seed 布局仍可接受。精确继承 cut 与逻辑 header 分离,并在读取正文后可用。
`session/end-seed` 加入了落盘词汇表。当前格式要求由 Session 拥有的已校验 marker 语义;冻结的 v0 codec 与迁移边负责哪些历史 v0 seed 布局仍可接受。精确继承 cut 与逻辑 header 分离,并在读取正文后可用。
[排队手动压缩决策](../feature/2026-07-30-queued-manual-compaction.zh.md)如今提供了第一个消费方。其尾部扫描会分别查找未匹配的 `compaction/start` 与最新 end-seed,只把位于该边界之后的 start 视为存活,并在同一个回放转换上清除不变量追踪状态。该谓词仍位于压缩功能所在的包中,不会成为通用核心辅助函数。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-31-goal-owned-durable-events.md
2026-07-31-goal-owned-durable-events.md: 8f7d3b74a83e063fbfc5a530f9e51a2a3eea8197
2026-07-31-goal-owned-durable-events.zh.md: 8ee0099209c9c98a88549073b33c98ddb4b12bbd
2026-07-31-goal-owned-durable-events.md: 0d3e632ba275f38e48fdf1f9ee6747f4bc460aab
2026-07-31-goal-owned-durable-events.zh.md: fd29a6f36eed42f5ba26a46a32acb165d1e42873
@@ -30,4 +30,4 @@ The domain does not automatically project each mutation into model input. Goal t
Goal state is independent of inbox placement and admission. Replay has one mutation path, projections advance directly on `goal/change`, and continuation messages carry only round attribution. The model does not receive a mutation-only `<goal_state>` message; model-visible state appears through goal tools and scheduled continuation prompts. Direct session writers remain trusted and can append malformed changes, which the strict fold and invariant companion reject.
Focused goal, goal-round-driver, command, TUI, and client-fixture tests pin durable replay, positive-round accounting, inbox independence, projection updates, and restored-session behavior. The keyless process test inspects the persisted `goal/change` event and verifies that creation alone starts no continuation round.
Focused goal, goal-round-driver, command, TUI, and Client RemoteMock tests pin durable replay, positive-round accounting, inbox independence, projection updates, and restored-session behavior. The keyless process test inspects the persisted `goal/change` event and verifies that creation alone starts no continuation round.
@@ -30,4 +30,4 @@ Goal 领域需要持久状态,但不需要拥有待处理的模型输入。继
Goal 状态不依赖 inbox 放置与准入。回放只有一条变更路径,投影直接由 `goal/change` 推进,继续执行消息只携带 Round 归属。模型不会收到仅用于变更的 `<goal_state>` 消息;模型可见状态来自 goal 工具与已调度的继续执行提示词。直接写入会话的写入方仍受信任,并且可以追加畸形变更;严格折叠与 invariant 配套模块会拒绝这些变更。
聚焦的 goal、goal-round-driver、command、TUI 与 client fixture(测试前置数据)测试固定持久回放、正数 Round 计数、inbox 独立性、投影更新和恢复会话行为。无密钥进程测试检查持久的 `goal/change` 事件,并验证仅创建 goal 不会启动继续执行 Round。
聚焦的 goal、goal-round-driver、command、TUI 与 Client RemoteMock 测试固定持久回放、正数 Round 计数、inbox 独立性、投影更新和恢复会话行为。无密钥进程测试检查持久的 `goal/change` 事件,并验证仅创建 goal 不会启动继续执行 Round。
@@ -1,6 +1,6 @@
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/simplification/2026-09-06-agent-request-freeze-provenance.md
2026-09-06-agent-request-freeze-provenance.md: 7a4816df61f6490647aba6f0603719e1b4662a20
2026-09-06-agent-request-freeze-provenance.zh.md: 239d7e69df1596010ef0f3c8789250f654a75cb1
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-07-31-ptc-runtime-python-fd3-protocol.md
2026-07-31-ptc-runtime-python-fd3-protocol.md: 097bcfd55f333a0973084f378a6c97cb630a4bec
2026-07-31-ptc-runtime-python-fd3-protocol.zh.md: ac656e421bd822994c50c33d0dfaf65ae46aa817
@@ -1,16 +1,16 @@
# Agent Note: the code-runtime-python fd-3 frame protocol
# Agent Note: the ptc-runtime-python fd-3 frame protocol
Status: implemented
The CPython code runtime now lives at `packages/experimental/code-runtime-python` (private, npm name `@deepseek-ai/dsh-experimental-code-runtime-python`); promotion to a released package follows the experimental-packages decision.
The CPython PTC runtime lives at `packages/experimental/ptc-runtime-python` and publishes as `@deepseek-ai/dsh-experimental-ptc-runtime-python`; the [publication decision](../process/2026-09-12-publish-all-experimental-packages.md) preserves its experimental status.
English | [中文](2026-07-31-code-runtime-python-fd3-protocol.zh.md)
English | [中文](2026-07-31-ptc-runtime-python-fd3-protocol.zh.md)
## Problem
`@deepseek-ai/dsh-experimental-code-runtime-python` owns the wire protocol intended for a CPython code-runtime provider. Such a provider runs each model program in a fresh `python3 -I` subprocess and bridges binding calls and completion values over the child's fd 3. The host cannot trust that channel: model code has full access to fd 3 and can forge any frame, so every inbound frame is hostile input that the host must validate and rebuild before reading. The protocol also has to carry lossless JSON without the depth limit `JSON.stringify` and `json.dumps` impose, because the seam's `CodeJsonValue` is depth-unbounded.
`@deepseek-ai/dsh-experimental-ptc-runtime-python` owns the wire protocol intended for a CPython ptc-runtime provider. Such a provider runs each model program in a fresh `python3 -I` subprocess and bridges binding calls and completion values over the child's fd 3. The host cannot trust that channel: model code has full access to fd 3 and can forge any frame, so every inbound frame is hostile input that the host must validate and rebuild before reading. The protocol also has to carry lossless JSON without the depth limit `JSON.stringify` and `json.dumps` impose, because the seam's `PtcJsonValue` is depth-unbounded.
The private experimental package contains both the protocol and runtime implementation: `PythonCodeRuntime` (the plugin's default export), the `python3 -I` subprocess path, and the Python-side JSON codec all live in `@deepseek-ai/dsh-experimental-code-runtime-python`. The protocol builds on the [portable identifier seam](../../archived/architecture/2026-07-31-code-runtime-portable-identifier-seam.md).
The experimental package contains both the protocol and runtime implementation: `PythonPtcRuntime` (the plugin's default export), the `python3 -I` subprocess path, and the Python-side JSON codec all live in `@deepseek-ai/dsh-experimental-ptc-runtime-python`. The protocol builds on the [portable identifier seam](../../archived/architecture/2026-07-31-code-runtime-portable-identifier-seam.md).
## Decision
@@ -42,4 +42,4 @@ Frames are JSON-lines on fd 3, one object per line, leaving stdout/stderr free f
Bought: the fd-3 protocol and its hostile-input codec form a self-contained, fully unit-covered layer, with an executing guard against TypeScript/Python field-set drift. The runtime built on it (`bootstrap.py`) consumes the reviewed wire contract.
Cost: the package name denotes a Python runtime family and `src/index.ts` exports the full `PythonCodeRuntime` implementation, so the protocol vocabulary is only one part of the package surface. The mirror e2e compares field names and required/optional status across the two sides but not field types; comparing type declarations across TypeScript and Python has no mechanical equivalent, so review and the runtime's real-subprocess suite retain that responsibility.
Cost: the package name denotes a Python runtime family and `src/index.ts` exports the full `PythonPtcRuntime` implementation, so the protocol vocabulary is only one part of the package surface. The mirror e2e compares field names and required/optional status across the two sides but not field types; comparing type declarations across TypeScript and Python has no mechanical equivalent, so review and the runtime's real-subprocess suite retain that responsibility.
@@ -1,16 +1,16 @@
# Agent Note: the code-runtime-python fd-3 frame protocol
# Agent Note: the ptc-runtime-python fd-3 frame protocol
Status: implemented
CPython 代码运行时现在位于 `packages/experimental/code-runtime-python`(私有,npm 名 `@deepseek-ai/dsh-experimental-code-runtime-python`);提升为发布包遵循 experimental-packages 决策。
CPython PTC 运行时位于 `packages/experimental/ptc-runtime-python`,以 `@deepseek-ai/dsh-experimental-ptc-runtime-python` 名称发布;[发布决策](../process/2026-09-12-publish-all-experimental-packages.zh.md)保留其实验性状态。
[English](2026-07-31-code-runtime-python-fd3-protocol.md) | 中文
[English](2026-07-31-ptc-runtime-python-fd3-protocol.md) | 中文
## Problem
`@deepseek-ai/dsh-experimental-code-runtime-python` 负责供 CPython code-runtime 提供方使用的 wire protocol。这样的提供方会在全新的 `python3 -I` 子进程中运行每个模型程序,并通过子进程 fd 3 桥接 binding 调用与完成值。Host 不能信任这条通道:模型代码可以完全访问 fd 3 并伪造任意帧,因此 host 必须把每个入站帧视为敌意输入,先校验并重建后才能读取。协议还必须承载无深度限制的 lossless JSON,因为 seam 的 `CodeJsonValue` 深度无界,而 `JSON.stringify` 和 `json.dumps` 都有递归深度限制。
`@deepseek-ai/dsh-experimental-ptc-runtime-python` 负责供 CPython ptc-runtime 提供方使用的 wire protocol。这样的提供方会在全新的 `python3 -I` 子进程中运行每个模型程序,并通过子进程 fd 3 桥接 binding 调用与完成值。Host 不能信任这条通道:模型代码可以完全访问 fd 3 并伪造任意帧,因此 host 必须把每个入站帧视为敌意输入,先校验并重建后才能读取。协议还必须承载无深度限制的 lossless JSON,因为 seam 的 `PtcJsonValue` 深度无界,而 `JSON.stringify` 和 `json.dumps` 都有递归深度限制。
这个私有实验包同时包含协议与 runtime 实现:`PythonCodeRuntime`(插件的默认导出)、`python3 -I` 子进程路径与 Python 侧 JSON codec 都在 `@deepseek-ai/dsh-experimental-code-runtime-python` 中。协议建立在[可移植标识符 seam](../../archived/architecture/2026-07-31-code-runtime-portable-identifier-seam.md)之上。
这个实验包同时包含协议与 runtime 实现:`PythonPtcRuntime`(插件的默认导出)、`python3 -I` 子进程路径与 Python 侧 JSON codec 都在 `@deepseek-ai/dsh-experimental-ptc-runtime-python` 中。协议建立在[可移植标识符 seam](../../archived/architecture/2026-07-31-code-runtime-portable-identifier-seam.md)之上。
## Decision
@@ -42,4 +42,4 @@ CPython 代码运行时现在位于 `packages/experimental/code-runtime-python`
收获:fd-3 协议及其敌意输入 codec 构成自包含、unit 全覆盖的一层,并由执行中的 guard 防止 TypeScript/Python 字段集漂移。基于它构建的 runtime(`bootstrap.py`)消费经过评审的 wire contract。
代价:包名表示 Python runtime 家族,而 `src/index.ts` 导出完整的 `PythonCodeRuntime` 实现,协议 vocabulary 只是包表面的一部分。mirror e2e 会比较两侧字段名与必填/可选状态,但不比较字段类型;跨 TypeScript 与 Python 比较类型声明没有机械等价物,因此评审与 runtime 的真实子进程套件继续负责这项检查。
代价:包名表示 Python runtime 家族,而 `src/index.ts` 导出完整的 `PythonPtcRuntime` 实现,协议 vocabulary 只是包表面的一部分。mirror e2e 会比较两侧字段名与必填/可选状态,但不比较字段类型;跨 TypeScript 与 Python 比较类型声明没有机械等价物,因此评审与 runtime 的真实子进程套件继续负责这项检查。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-03-per-session-agent-presets.md
2026-08-03-per-session-agent-presets.md: 14d568689a662e3e7d1fedf26c22aea6faddd56b
2026-08-03-per-session-agent-presets.zh.md: 9b296fa7a1c2b9164d9e7a0ad676929179a64fef
2026-08-03-per-session-agent-presets.md: 47f2e28c2aac583ce4ca810157e9c7abe58256d5
2026-08-03-per-session-agent-presets.zh.md: 88b643ea710364f18ff7c73a8ca0e10de0fbb44c
@@ -27,13 +27,13 @@ The presets the deployment ships are the directories under `packages/preset/agen
Mounting is per-session by default. Measured cost for a twelve-row composition is ~3ms and ~600KB per session, so isolation is the cheaper default than any sharing scheme, and a preset authored by a user or by an agent then has the smallest possible blast radius. A preset that genuinely owns an expensive singleton opts into sharing with Cordis's own `isolate` vocabulary: a named realm label is process-global, so two subtrees naming the same label resolve one instance.
Which preset an unnamed session gets is a user setting (`agent-presets.default`) layered over the composition's own `default`, which becomes the `base`. Both layers are needed: the composition value is what a deployment ships and must keep working with no settings provider at all, and the setting is what a person changes without editing a `cordis.yml` they may not own.
The `agent-presets` user-settings namespace carries `modeSelectionEnabled` and `default`. `modeSelectionEnabled` defaults to `true`: the existing new-session picker remains present and an unnamed session resolves to the saved user `default`, or the composition's deployment `default` when none exists. The Web Settings toggle changes only that policy: disabling selection temporarily uses the deployment default, while re-enabling it restores the saved user `default`. This is a deliberate exception to the ordinary user-over-composition settings precedence established in #1539: hiding the chooser disables the user's mode-selection policy without deleting its saved value. The Host policy governs every later session whose caller omits a preset; explicitly named presets and existing sessions remain unchanged. The composition value also keeps the package working with no settings provider, while an enabled user override changes later sessions without editing a deployment-owned `cordis.yml`.
## Consequences
**The effective default is read per resolution, never snapshotted.** A cached value would need a `watch` subscription and a reload path to stay honest, and the resolved scope already re-reads a hot-reloaded document. Reading through is also what makes the boundary correct rather than merely cheap: the new value applies to the next session created, and every running session keeps the composition it was built from. That invariant is the same one the session log enforces from the other side — the header records the id a session was CREATED with and an `agent-preset/selected` event records any later blank-session switch, so a reader resolves the pair (`resolveSessionPreset`) and never the header alone: a resume rebuilds the composition its history was produced under rather than the deployment default at resume time, a cold transcript's presenters resolve in that composition's layer, and the gateway rejects an attempt to adopt a live session under a preset other than the one it currently runs. A snapshot would make the two disagree at exactly the moment the setting changes.
**The effective default is read per resolution, never snapshotted.** A cached value would need a `watch` subscription and a reload path to stay honest, and the resolved scope already re-reads a hot-reloaded document. The Host setting itself applies when an unnamed session is resolved afterwards. An explicit Web Settings action additionally routes its accepted effective default through the existing blank-session selection path only when the captured session id is still current and blank; it never recomposes a running session or rewrites that session's history. The session log enforces the same invariant from the other side — the header records the id a session was CREATED with and an `agent-preset/selected` event records any later blank-session switch, so a reader resolves the pair (`resolveSessionPreset`) and never the header alone: a resume rebuilds the composition its history was produced under rather than the deployment default at resume time, a cold transcript's presenters resolve in that composition's layer, and the gateway rejects an attempt to adopt a live session under a preset other than the one it currently runs. A snapshot would make the two disagree at exactly the moment the setting changes.
**A directly-plugged subtree is invisible to the boot audit.** It never links itself to an `Entry`, so it is absent from `ctx.loader.entries()` and `assertEntriesActivated` cannot see it. The mount audits its own rows instead, reading the tree through an `Include` subclass that publishes it.
**A directly-plugged subtree is invisible to the boot audit.** It never links itself to an `Entry`, so it is absent from `ctx.loader.entries()` and `auditStartupEntries` cannot see it. The mount reads its own rows through an `Include` subclass that publishes the tree and requires every enabled row to activate, independently of the [application startup policy](2026-09-09-consumer-owned-startup-strictness.md).
**A preset can only name a group because the app registers one.** Sharing a realm across rows is a `cordis:group` row, and a preset living outside this workspace — the authored ones under the Harness home, which is the point — cannot resolve `@cordisjs/plugin-group` by name: Node's upward `node_modules` walk never reaches the harness from there. `boot()` therefore registers `cordis:group` beside `cordis:include` as a loader builtin, so both load through the ambient module pipeline rather than through the included tree's own specifier resolution. Without it the `isolate` vocabulary above is expressible one row at a time only, and a provider could never be grouped with its consumers.
@@ -61,11 +61,11 @@ Which preset an unnamed session gets is a user setting (`agent-presets.default`)
**A preset's package names must resolve from the harness, not from the preset.** `EntryTree.import()` resolves a row against its own tree's `baseUrl`, which `Include` sets to the composition's directory. That is right for a relative specifier and fatal for a package name: a locally authored preset lives under the user's home, where Node's upward `node_modules` walk never reaches the installed harness, so every `@deepseek-ai/dsh-*` row fails to import and the whole preset is unmountable. The shipped presets hid this — they sit inside the install. The mount records the host composition's base before plugging the subtree and sends bare specifiers there, leaving relative paths resolving from the preset so its own files still travel with it. The real-composition test writing a preset into a temp root is what found it.
**The preset id is model-visible and must be logged.** It determines the tool set and prompt, so a resumed session has to restore the same composition; recording it is a session fact, not runtime state. It rides the session header beside `cwd`, and the summary carries it so a picker shows what a session actually runs rather than the deployment's current default.
**The preset id is model-visible and must be logged.** It determines the tool set and prompt, so a resumed session has to restore the same composition; recording it is a session fact, not runtime state. It rides the session header beside `cwd`, and the summary carries it so the client surfaces show what a session actually runs rather than the deployment's current default.
**A durable header field is not durable until the provider writes it.** `agentPreset` landed on `SessionHeader` with the right rationale and the JSONL provider omitted it; the derived query index also maps header fields explicitly, so a resumed Session came back with no preset and the surfaces that name it fell silent. `summarizeCold` had the same form — it hand-built the cold list row instead of reusing the shared projection. A field declared durable needs a test that crosses a real store, not only the type that declares it.
**The choice belongs to the screen where it still works.** The composer seat spent almost its whole life disabled, since the preset is fixed once a turn has run. It moved to the new-session screen beside the workspace picker, where the pick is *staged*: that screen precedes the session it applies to, and the stage lands when a session becomes current and is still blank — covering both the session a workspace connect creates and the blank one it reuses, which riding `sessions.create` would miss. It is spent on first use, matching the workspace picker beside it. What a running session runs is then a read-only label in its header: a control there would promise a switch the host refuses outright.
**The choice belongs to the screen where it still works.** The control lives on the new-session screen beside the workspace picker, is present under the default-on Host policy, and disappears only after `modeSelectionEnabled` is disabled. Its pick is *staged*: that screen precedes the session it applies to, and the stage lands when a session becomes current and is still blank — covering both the session a workspace connect creates and the blank one it reuses, which riding `sessions.create` would miss. It is spent on first use, matching the workspace picker beside it; hiding the picker discards a stage that has not reached a session and returns the current blank session to the deployment default through the same selection path. What a running or historical session runs remains a read-only label in its header: a control there would promise a switch the host refuses outright.
**A preset multiplies a cost the host was already paying: nothing disposes an agent.** Measured against the shipped compositions with `--expose-gc`, one live agent holds ~0.17 MB on `minimal` and ~1.31 MB on `standard`/`cordis`, mounting in ~38 ms and ~135 ms; the first agent of a process costs ~7 MB more as Node imports the modules, which every later mount then shares. Growth is strictly linear — 10, 30 and 50 agents give the same per-agent delta — and disposal reclaims essentially all of it (50 `standard` agents held 57.8 MB and returned it). So the object graph does not leak; the lifecycle does. `ApiSessionAgentController` discards the `AgentHandle` returned by the registry, `archiveSession` only edits the workspace registry, `AgentRegistry` has no eviction, and the sole disposal site in the host is the JSON-RPC server's own shutdown. A web host therefore retains every session it has touched, at ~1.3 MB each once presets are composed rather than ~0.2 MB before. Note that pruning the mount registry does not help here: it drops records whose fiber `uid` has cleared, and an agent that never dies never clears one.
@@ -27,14 +27,13 @@ Status: implemented
挂载默认按会话进行。实测一份十二行组装每会话约 3ms、约 600KB,因此隔离比任何共享方案都更划算;而由用户或 agent 写出的 preset 也因此拥有尽可能小的影响面。确实自带昂贵单例的 preset,可以用 Cordis 自身的 `isolate` 词汇显式选择共享:命名 realm 的 label 是进程级全局的,因此两棵子树只要写同一个 label 就解析到同一个实例。
未指名 preset 的会话拿到哪一个,是一项用户设置(`agent-presets.default`),叠在组装自身的 `default` 之上——后者成为 `base`。两层都需要:组装里的值是部署交付的东西,在完全没有 settings 提供方时也必须照常工作;而设置是让人不必去改一份可能并不属于自己的 `cordis.yml` 就能调整的东西。
`agent-presets` 用户设置命名空间同时携带 `modeSelectionEnabled` 与 `default`。`modeSelectionEnabled` 默认为 `true`:既有的新建会话选择器保持显示;未指名会话会解析到已保存的用户 `default`,尚未保存时则使用组装中 `default` 指定的部署默认值。Web 设置开关只改变该策略:关闭选择时临时使用部署默认值,再次开启时恢复已保存的用户 `default`。这是对 #1539 所确立“用户值覆盖组装值”这一普通 settings 优先级的有意例外:隐藏选择器会停用用户的模式选择策略,但不会删除其保存值。该 Host 策略适用于此后所有未显式指定 preset 的会话;显式指定及既有会话不受影响。组装值还使本包在没有 settings 提供方时照常工作;选择器开启后,用户可覆盖默认值来改变后续会话,而无需编辑部署所拥有的 `cordis.yml`。
## 后果
**有效默认值在每次解析时读取,绝不保存快照。** 缓存下来就需要一个 `watch` 订阅和一条重载路径才能保持诚实,而解析后的 scope 本来就会重读热重载过的文档。读穿也不只是省事,它让边界本身是对的:新值作用于**下一个新建的会话**,每个运行中的会话保持它被构建时的那份组装。这条不变量正是 session 日志从另一侧执行的同一条——header 记录会话**创建时**的 id,此后空白期的任何切换由 `agent-preset/selected` 事件记录,因此读取方解析的是两者之和(`resolveSessionPreset`)、绝不单看 header:恢复重建的是其历史所产出的那份组装而不是恢复时的部署默认值,冷读记录的 presenter 在那份组装的层里解析,网关也会拒绝把一个活着的会话收编到它当前运行的 preset 以外的 preset 之下。快照会让两者恰好在设置改变的那一刻各说各话。
**有效默认值在每次解析时读取,绝不保存快照。** 缓存下来就需要一个 `watch` 订阅和一条重载路径才能保持诚实,而解析后的 scope 本来就会重读热重载过的文档。Host 设置本身会在此后解析未指名会话时生效。Web Settings 中的明确操作还会把已接受的有效默认值送入既有的空白会话选择链路,但只在操作前捕获的会话 id 仍是当前空白会话时对齐;它绝不会重新组装运行中的会话,也不会改写该会话的历史。session 日志从另一侧执行同一条不变量——header 记录会话**创建时**的 id,此后空白期的任何切换由 `agent-preset/selected` 事件记录,因此读取方解析的是两者之和(`resolveSessionPreset`)、绝不单看 header:恢复重建的是其历史所产出的那份组装而不是恢复时的部署默认值,冷读记录的 presenter 在那份组装的层里解析,网关也会拒绝把一个活着的会话收编到它当前运行的 preset 以外的 preset 之下。快照会让两者恰好在设置改变的那一刻各说各话。
**直接挂载的子树对启动审计不可见。** 它不会把自己关联到 `Entry`,因此不在 `ctx.loader.entries()` 中,`assertEntriesActivated` 也看不到它。改由挂载过程自行校验各行,通过一个会公开自身 tree 的 `Include` 子类读取。
**直接挂载的子树对启动审计不可见。** 它不会把自己关联到 `Entry`,因此不在 `ctx.loader.entries()` 中,`auditStartupEntries` 也看不到它。挂载过程通过一个会公开自身 tree 的 `Include` 子类读取各行,并要求每个启用行都激活,不受[应用启动策略](2026-09-09-consumer-owned-startup-strictness.zh.md)影响。
**preset 能写出 group,是因为 app 注册了它。** 跨行共享 realm 就是一个 `cordis:group` 行,而住在本工作区之外的 preset——也就是 Harness home 下由人或 agent 创作的那些,正是这套设计的目的——无法按名字解析 `@cordisjs/plugin-group`:Node 向上查找 `node_modules` 的路径从那里永远走不到 harness。因此 `boot()` 把 `cordis:group` 与 `cordis:include` 并排注册为 loader builtin,两者都经由环境模块管线加载,而不依赖被包含树自身的说明符解析。没有它,上文那套 `isolate` 词汇就只能一行一行地表达,提供方也永远无法与它的消费方归入同一组。
@@ -62,11 +61,11 @@ Status: implemented
**preset 的包名必须从 harness 解析,而非从 preset 解析。** `EntryTree.import()` 按行所属树的 `baseUrl` 解析,而 `Include` 把它设为组装文件所在的目录。这对相对标识符是对的,对包名却是致命的:本地创作的 preset 位于用户主目录之下,Node 向上查找 `node_modules` 永远够不到已安装的 harness,因此每一个 `@deepseek-ai/dsh-*` 行都会导入失败,整个 preset 无法挂载。随部署提供的 preset 掩盖了这一点——它们本就在安装目录之内。挂载在插入子树之前先记录宿主组装的基址,并把裸标识符送往那里,同时让相对路径继续从 preset 解析,使它自带的文件仍随它一同迁移。发现它的正是那个把 preset 写入临时根目录的真实组装测试。
**preset id 对模型可见,必须写入日志。** 它决定工具集与提示词,因此被恢复的会话必须还原同一份组装;记录它属于会话事实,而非运行时状态。它与 `cwd` 并列写在会话头部,并由会话摘要携带,使选择器显示的是某个会话实际运行的 preset,而非部署当前的默认值。
**preset id 对模型可见,必须写入日志。** 它决定工具集与提示词,因此被恢复的会话必须还原同一份组装;记录它属于会话事实,而非运行时状态。它与 `cwd` 并列写在会话头部,并由会话摘要携带,使客户端界面显示某个会话实际运行的 preset,而非部署当前的默认值。
**持久化 header 字段在 provider 写入前都算不上持久。** `agentPreset` 带着正确理由落在 `SessionHeader` 上,而 JSONL provider 遗漏了它;派生 query index 也显式映射 header 字段,于是恢复后的 Session 没有 preset,所有据以命名它的 surface 随之失声。`summarizeCold` 是同一种形式——它手工拼装 cold list row,而没有复用共享 projection。声明为持久的字段,需要一个跨越真实 store 的测试,而不只是声明它的类型。
**这个选择属于它仍然可用的那个界面。** composer 座位几乎一生都处于禁用状态,因为一旦跑过一个轮次,preset 即固定。它移到了新建会话界面、工作区选择器旁边,选择在那里是**暂存**的:该界面先于它要应用到的会话存在,暂存值在某个会话成为当前会话且仍为空白时落地——这既覆盖工作区连接新建的会话,也覆盖它复用的那个空白会话,而搭 `sessions.create` 的便车会漏掉后者。它一经使用即被清空,与旁边的工作区选择器一致。至于运行中的会话在跑什么,则是其标题旁的一个只读标签:在那里放控件,等于承诺一次宿主会断然拒绝的切换。
**这个选择属于它仍然可用的那个界面。** 控件位于新建会话界面、工作区选择器旁边,在 Host 默认开启策略下直接显示,仅在 `modeSelectionEnabled` 关闭后隐藏。选择在那里是**暂存**的:该界面先于它要应用到的会话存在,暂存值在某个会话成为当前会话且仍为空白时落地——这既覆盖工作区连接新建的会话,也覆盖它复用的那个空白会话,而搭 `sessions.create` 的便车会漏掉后者。它一经使用即被清空,与旁边的工作区选择器一致;隐藏选择器会丢弃尚未到达会话的暂存选择,并通过同一条选择链路把当前空白会话带回部署默认值。至于运行中或历史会话在跑什么,仍由其标题旁的只读标签展示:在那里放控件,等于承诺一次宿主会断然拒绝的切换。
**preset 放大的是宿主本来就在付的代价:没有任何东西会 dispose 一个 agent。** 用 `--expose-gc` 对随附组装实测:一个存活的 agent 在 `minimal` 上约占 0.17 MB、在 `standard`/`cordis` 上约 1.31 MB,挂载耗时分别约 38 ms 与 135 ms;进程里第一个 agent 另需约 7 MB,那是 Node 首次 import 模块的一次性成本,此后每次挂载共享。增长严格线性——10、30、50 个的单个增量一致——且 dispose 后基本全额回收(50 个 `standard` 占住 57.8 MB,释放后全部归还)。所以对象图并不泄漏,缺的是生命周期。`ApiSessionAgentController` 会丢弃注册表返回的 `AgentHandle`,`archiveSession` 只改工作区注册表,`AgentRegistry` 没有驱逐机制,而宿主里唯一一处 dispose 是 JSON-RPC 服务器自身的关停。于是一个 web 宿主会留住它接触过的每一个会话,组装 preset 之后每个约 1.3 MB,而在此之前约 0.2 MB。注意:剪枝挂载注册表在这里没有用——它丢弃的是 fiber `uid` 已清空的记录,而永不死亡的 agent 永远不会清空它。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-04-draft-provider-endpoint-interrogation.md
2026-08-04-draft-provider-endpoint-interrogation.md: 462b8ba3e771018126aa6483a8cd72999361144d
2026-08-04-draft-provider-endpoint-interrogation.zh.md: c7612513b95bc3dc9736b9bfaf70cca26a776181
2026-08-04-draft-provider-endpoint-interrogation.md: 469ab38b0f5609808a9df03237d4d21c0080d843
2026-08-04-draft-provider-endpoint-interrogation.zh.md: 74c6bc7ed73572d5e99ea973c9ea94d22319cbdf
@@ -17,7 +17,7 @@ The awkward part is that the question is about something that does not exist yet
Interrogation is keyed by **settings namespace**, not by provider route:
- `ctx.llm.registerModelDiscovery(settingsNs, discover)` lets an adapter plugin offer to interrogate endpoints for the namespace it owns, and `ctx.llm.discoverModels(settingsNs, request)` asks. There is no way to enumerate which namespaces registered: a surface that cannot interrogate learns it from the refusal, and a list nothing consumed would be a required wire field doing nothing. The namespace is the right key because a configuration surface already holds it from the configurable-provider directory, and because a provider being added has no route to name.
- `LlmModelDiscoveryRequest` carries the draft — an optional `provider`, an optional `baseURL`, an optional `api`, an optional `apiKey`, and a signal — and needs at least one of `provider` or `baseURL` to have anything to answer about. `provider` exists because a route the adapter already describes is answered from its own registry with no network call at all; only a route it does not describe reaches an endpoint. Nothing in this path writes settings or credentials. A named configured route reads its stored credential and deployment-owned profile `headers` inside the Host: the credential is write-only and the curated Models page does not edit headers, so neither can be reconstructed from that page's draft. The typed key wins over the stored credential, while the profile headers still accompany the request.
- `LlmModelDiscoveryRequest` carries the draft — an optional `provider`, an optional `baseURL`, an optional `api`, an optional `apiKey`, and a signal — and needs at least one of `provider` or `baseURL` to have anything to answer about. `provider` exists because a route the adapter already describes is answered from its own registry with no network call at all; only a route it does not describe reaches an endpoint. The custom-provider card parses a non-empty `baseURL` and permits only HTTP or HTTPS before it calls this operation or writes the profile. Nothing in this path writes settings or credentials. A named configured route reads its stored credential and deployment-owned profile `headers` inside the Host: the credential is write-only and the curated Models page does not edit headers, so neither can be reconstructed from that page's draft. The typed key wins over the stored credential, while the profile headers still accompany the request.
- `LlmDiscoveredModel` makes every field but `id` optional, because most listings disclose an id and nothing else. The reply is candidates, not a catalog: a surface adopting one still owes the capacities the adapter requires.
- `llm.discoverModels` carries the same draft over the wire. Its `apiKey` is the third and last payload on which a secret may ride, alongside `settings.update`/`mutate` and `credentials.set`, and it is never stored or echoed back. It does ride the client's outgoing envelope like every other secret-bearing payload, where a `subscribeEnvelopes()` observer can see it; redacting that tap is a configuration-plane-wide change, not this method's to make alone. Connection authenticates the method with the complete Host API: it makes the host issue a GET to a caller-chosen URL and reports the outcome, which an anonymous caller must not receive. Every refusal folds into `model-discovery-failed`, whose message is the adapter's own text and whose details name the endpoint asked but never the credential offered.
@@ -41,10 +41,10 @@ pi-ai supplies `createProvider({ fetchModels })` plus `Models.refresh()` and a `
## Consequences
A person adding a gateway can ask it what it serves instead of hunting through its documentation, and the answer arrives as candidates they choose from rather than as configuration written behind their back. When an endpoint discloses richer metadata, adopting a candidate fills its id, name, context window, and output-token cap into the editable Web row. Search preserves hidden selections, selecting all adds the visible results, and deselecting all clears every result so a filtered picker cannot submit hidden models accidentally. An already-configured enterprise gateway uses the same deployment headers and Harness `User-Agent` for interrogation and model requests without adding a header injection field to the browser protocol. The seam gained a registry that is deliberately small: one offer per namespace, no storage, no lifecycle beyond the fiber.
A person adding a gateway can ask it what it serves instead of hunting through its documentation, and the answer arrives as candidates they choose from rather than as configuration written behind their back. A missing scheme, unsupported scheme, or malformed URL fails on the field without a Remote call; localhost, IPv4 and IPv6 literals, and custom ports remain valid. Provider and network failures remain distinct interrogation results. When an endpoint discloses richer metadata, adopting a candidate fills its id, name, context window, and output-token cap into the editable Web row. Search preserves hidden selections, selecting all adds the visible results, and deselecting all clears every result so a filtered picker cannot submit hidden models accidentally. An already-configured enterprise gateway uses the same deployment headers and Harness `User-Agent` for interrogation and model requests without adding a header injection field to the browser protocol. The seam gained a registry that is deliberately small: one offer per namespace, no storage, no lifecycle beyond the fiber.
What it costs: the wire gained a third secret-carrying payload, so the configuration plane's write-only surface is now three methods rather than two. Discovery coverage remains protocol-shaped rather than provider-shaped, and an endpoint using an unsupported request contract must be filled in by hand. Because nothing re-runs the question, a model list is still only as current as its last edit; that is the same trade the layer below made deliberately.
## Testing
`packages/llm/llm/tests/topology.spec.ts` covers the registry: one offer per namespace, disposal with the fiber, normalization that drops duplicate and unusable ids without inventing capacities, the `NO_DISCOVERY`/`INVALID_DISCOVERY` refusals, and the `model-discovery-failed` Remote mapping. `packages/llm/llm-pi-ai/tests/discovery.spec.ts` drives the probe against local HTTP servers — standard arrays and enriched objects with every accepted metadata spelling, Anthropic's native path, headers, and capacity fields, route keys that differ from nested canonical ids, name fallback, a preserved deployment path, an absent credential, a configured route supplying its stored credential and headers while a typed key wins without resolving the stored one, a catalog route answering without resolving one at all, dropped rows, 401/403 versus a server fault, a non-listing and a non-JSON body, an unreachable endpoint, caller cancellation, an unsupported protocol, and the size ceiling in both its declared-length and streamed forms. `packages/llm/llm-pi-ai/tests/loader-composition.spec.ts` boots settings and credentials through the Loader and proves settings-only headers reach `GET /models` with request-owned headers winning collisions. `packages/llm/llm-pi-ai/tests/adapter.spec.ts` rejects profile headers Fetch cannot represent, and `packages/llm/llm-pi-ai/tests/dynamic-config.spec.ts` proves a settings write reports that configuration error while its last good routes keep serving. `packages/client/connection/tests/node-half.host.spec.ts` pins the `llm/discoverModels` `/api` carrier registration, while the component and built-Web settings tests verify that the complete draft reaches the Remote, absent fields stay absent, selected metadata fills all four editable model fields, tuned rows win over rediscovery, filtered deselection clears hidden candidates, and no settings namespace or credential is written before selection.
`packages/llm/llm/tests/topology.spec.ts` covers the registry: one offer per namespace, disposal with the fiber, normalization that drops duplicate and unusable ids without inventing capacities, the `NO_DISCOVERY`/`INVALID_DISCOVERY` refusals, and the `model-discovery-failed` Remote mapping. `packages/llm/llm-pi-ai/tests/discovery.spec.ts` drives the probe against local HTTP servers — standard arrays and enriched objects with every accepted metadata spelling, Anthropic's native path, headers, and capacity fields, route keys that differ from nested canonical ids, name fallback, a preserved deployment path, an absent credential, a configured route supplying its stored credential and headers while a typed key wins without resolving the stored one, a catalog route answering without resolving one at all, dropped rows, 401/403 versus a server fault, a non-listing and a non-JSON body, an unreachable endpoint, caller cancellation, an unsupported protocol, and the size ceiling in both its declared-length and streamed forms. `packages/llm/llm-pi-ai/tests/loader-composition.spec.ts` boots settings and credentials through the Loader and proves settings-only headers reach `GET /models` with request-owned headers winning collisions. `packages/llm/llm-pi-ai/tests/adapter.spec.ts` rejects profile headers Fetch cannot represent, and `packages/llm/llm-pi-ai/tests/dynamic-config.spec.ts` proves a settings write reports that configuration error while its last good routes keep serving. `packages/client/connection/tests/node-half.host.spec.ts` pins the `llm/discoverModels` `/api` carrier registration, while the component and built-Web settings tests verify that the complete draft reaches the Remote, malformed and non-HTTP(S) custom-provider URLs make no request, localhost and IP literals with custom ports remain valid, provider failures stay distinct from URL errors, absent fields stay absent, selected metadata fills all four editable model fields, tuned rows win over rediscovery, filtered deselection clears hidden candidates, and no settings namespace or credential is written before selection.
@@ -17,7 +17,7 @@ Status: implemented
询问以 **settings namespace** 为键,而不是提供方路由:
- `ctx.llm.registerModelDiscovery(settingsNs, discover)` 让适配器插件为自己拥有的 namespace 提供「询问端点」的能力,`ctx.llm.discoverModels(settingsNs, request)` 发起询问。没有任何办法枚举哪些 namespace 注册过:询问不了的界面会从那句拒绝里知道,而一份无人消费的列表只会变成一个什么都不做的必填协议字段。以 namespace 为键是对的,因为配置界面已经从可配置提供方目录里拿到了它,也因为正在新增的提供方没有路由可点名。
- `LlmModelDiscoveryRequest` 携带草稿——可选的 `provider`、可选的 `baseURL`、可选的 `api`、可选的 `apiKey`,以及一个 signal——且 `provider` 与 `baseURL` 至少要有一个,才有东西可答。`provider` 之所以存在,是因为适配器已经描述过的路由直接由它自己的注册表作答、完全不联网;只有它未描述的路由才会抵达某个端点。这条路径不写 settings 与 credentials。已配置且具名的路由会在 Host 内读取已存凭据和部署方持有的 profile `headers`:凭据只写,而精选的 Models 页面不编辑 headers,因此页面草稿无法重建两者。键入的密钥优先于已存凭据,profile headers 则仍随请求发送。
- `LlmModelDiscoveryRequest` 携带草稿——可选的 `provider`、可选的 `baseURL`、可选的 `api`、可选的 `apiKey`,以及一个 signal——且 `provider` 与 `baseURL` 至少要有一个,才有东西可答。`provider` 之所以存在,是因为适配器已经描述过的路由直接由它自己的注册表作答、完全不联网;只有它未描述的路由才会抵达某个端点。自定义提供方卡片会解析非空的 `baseURL`,且只允许 HTTP 或 HTTPS,之后才会调用该操作或写入 profile。这条路径不写 settings 与 credentials。已配置且具名的路由会在 Host 内读取已存凭据和部署方持有的 profile `headers`:凭据只写,而精选的 Models 页面不编辑 headers,因此页面草稿无法重建两者。键入的密钥优先于已存凭据,profile headers 则仍随请求发送。
- `LlmDiscoveredModel` 除 `id` 外每个字段都可选,因为大多数列表只公布 id。回复是候选而非 catalog:采纳其中一条的界面仍要补上适配器所需的容量。
- `llm.discoverModels` 把同一份草稿送过协议层。它的 `apiKey` 是可承载机密的第三个、也是最后一个载荷(另两个是 `settings.update`/`mutate` 与 `credentials.set`),且绝不被存储或回显。它确实会像其他承载机密的载荷一样随客户端外发信封同行,`subscribeEnvelopes()` 观察者看得到;把那个抽头脱敏是整个配置面的改动,不该由这一个方法独自决定。Connection 用与完整 Host API 相同的会话认证该方法:它让宿主向调用方选定的 URL 发起 GET 并回报结果,匿名调用者绝不能获得这类探测能力。每一种拒绝都折叠为 `model-discovery-failed`,其消息是适配器自己的文本,details 点名被询问的端点,绝不点名所提供的凭据。
@@ -41,10 +41,10 @@ pi-ai 提供了 `createProvider({ fetchModels })` 加上 `Models.refresh()` 与
## Consequences
接入网关的人可以直接问它服务什么,而不必去翻它的文档;答案以候选形式抵达,由用户自己挑选,而不是被背着写进配置。当端点公布了更丰富的元数据时,采纳候选会把 id、名称、上下文窗口与最大输出 token 数填进 Web 的可编辑行。搜索会保留隐藏项的勾选状态,全选会加入可见结果,而取消全选会清空全部结果,因此筛选后的选择器不会意外提交隐藏模型。已配置的企业网关会为询问与模型请求使用同一组部署 headers 和 Harness `User-Agent`,而无需给浏览器协议增加 header 注入字段。seam 因此多了一个刻意保持很小的注册表:每个 namespace 一份、不存储、生命周期不超出 fiber。
接入网关的人可以直接问它服务什么,而不必去翻它的文档;答案以候选形式抵达,由用户自己挑选,而不是被背着写进配置。缺少 scheme、不支持的 scheme 或 URL 格式错误都会直接在字段处失败,不会发起 Remote 调用;localhost、IPv4 与 IPv6 字面地址以及自定义端口仍然有效。提供方与网络失败仍是独立的询问结果。当端点公布了更丰富的元数据时,采纳候选会把 id、名称、上下文窗口与最大输出 token 数填进 Web 的可编辑行。搜索会保留隐藏项的勾选状态,全选会加入可见结果,而取消全选会清空全部结果,因此筛选后的选择器不会意外提交隐藏模型。已配置的企业网关会为询问与模型请求使用同一组部署 headers 和 Harness `User-Agent`,而无需给浏览器协议增加 header 注入字段。seam 因此多了一个刻意保持很小的注册表:每个 namespace 一份、不存储、生命周期不超出 fiber。
代价是:协议层多了第三个承载机密的载荷,配置面的只写接口从两个方法变成三个。发现覆盖范围仍按协议而非按提供方划分,使用不受支持请求约定的端点仍须手工填写。由于没有任何环节会重跑该询问,模型列表的新鲜度依旧只到最近一次编辑为止;这与下层刻意做出的取舍是同一个。
## Testing
`packages/llm/llm/tests/topology.spec.ts` 覆盖注册表:每个 namespace 一份、随 fiber dispose(资源释放)、丢弃重复与不可用 id 且不凭空补容量的归一化、`NO_DISCOVERY`/`INVALID_DISCOVERY` 两种拒绝,以及 `model-discovery-failed` Remote 映射。`packages/llm/llm-pi-ai/tests/discovery.spec.ts` 针对本地 HTTP 服务器驱动探测——包括采用每种受支持元数据拼写的标准数组与富信息对象、Anthropic 原生路径、headers 与容量字段、不同于嵌套规范 id 的路由键、名称回退、被保留的部署路径、无凭据、已配置路由提供自己的已存凭据与 headers 且键入的密钥无需解析已存凭据便可压过它、catalog 路由完全不解析凭据即作答、被丢弃的行、401/403 与服务器故障之别、非列表与非 JSON 响应、不可达端点、调用方取消、不支持的协议,以及尺寸上限的「声明长度」与「流式」两种形态。`packages/llm/llm-pi-ai/tests/loader-composition.spec.ts` 通过 Loader 启动 settings 与 credentials,并证明仅配置在 settings 中的 headers 会抵达 `GET /models`,且请求所持有的 headers 赢得冲突。`packages/llm/llm-pi-ai/tests/adapter.spec.ts` 拒绝 Fetch 无法表示的 profile headers,`packages/llm/llm-pi-ai/tests/dynamic-config.spec.ts` 证明 settings 写入会报告该配置错误,同时上一组可用路由仍继续服务。`packages/client/connection/tests/node-half.host.spec.ts` 固定 `llm/discoverModels` 的 `/api` 承载注册,而设置页的组件测试和构建后 Web 测试则验证完整草稿抵达 Remote、缺席字段保持缺席、所选元数据填满四个可编辑模型字段、用户调整过的行优先于重新发现结果、筛选后的取消选择会清除隐藏候选项,以及选择前没有 settings namespace 或凭据被写入。
`packages/llm/llm/tests/topology.spec.ts` 覆盖注册表:每个 namespace 一份、随 fiber dispose(资源释放)、丢弃重复与不可用 id 且不凭空补容量的归一化、`NO_DISCOVERY`/`INVALID_DISCOVERY` 两种拒绝,以及 `model-discovery-failed` Remote 映射。`packages/llm/llm-pi-ai/tests/discovery.spec.ts` 针对本地 HTTP 服务器驱动探测——包括采用每种受支持元数据拼写的标准数组与富信息对象、Anthropic 原生路径、headers 与容量字段、不同于嵌套规范 id 的路由键、名称回退、被保留的部署路径、无凭据、已配置路由提供自己的已存凭据与 headers 且键入的密钥无需解析已存凭据便可压过它、catalog 路由完全不解析凭据即作答、被丢弃的行、401/403 与服务器故障之别、非列表与非 JSON 响应、不可达端点、调用方取消、不支持的协议,以及尺寸上限的「声明长度」与「流式」两种形态。`packages/llm/llm-pi-ai/tests/loader-composition.spec.ts` 通过 Loader 启动 settings 与 credentials,并证明仅配置在 settings 中的 headers 会抵达 `GET /models`,且请求所持有的 headers 赢得冲突。`packages/llm/llm-pi-ai/tests/adapter.spec.ts` 拒绝 Fetch 无法表示的 profile headers,`packages/llm/llm-pi-ai/tests/dynamic-config.spec.ts` 证明 settings 写入会报告该配置错误,同时上一组可用路由仍继续服务。`packages/client/connection/tests/node-half.host.spec.ts` 固定 `llm/discoverModels` 的 `/api` 承载注册,而设置页的组件测试和构建后 Web 测试则验证完整草稿抵达 Remote、格式错误和非 HTTP(S) 的自定义提供方 URL 不会发起请求、带自定义端口的 localhost 与 IP 字面地址仍然有效、提供方失败与 URL 错误保持区别、缺席字段保持缺席、所选元数据填满四个可编辑模型字段、用户调整过的行优先于重新发现结果、筛选后的取消选择会清除隐藏候选项,以及选择前没有 settings namespace 或凭据被写入。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-05-profile-plugin-bundles.md
2026-08-05-profile-plugin-bundles.md: 7e51345e7eba8a58db63807e31d4a11481e3ffea
2026-08-05-profile-plugin-bundles.zh.md: b2631603737ea9412eb97029ff01d751d8084cec
2026-08-05-profile-plugin-bundles.md: 48786a9c1ccaceb5f16c9eefed01e556cf759e6d
2026-08-05-profile-plugin-bundles.zh.md: c88cbf98e4276549a6fae6a5b1253d3d838a64ca
@@ -14,7 +14,7 @@ Everything becomes a **profile**: a directory `$DSH_HOME/profiles/<name>` with a
The default Profile templates use `@deepseek-ai/dsh-base` as the shared core for `web`, `headless`, `sdk`, and `acp`, with one mode bundle above it. The [standalone `sdk-minimal` profile](../../../../packages/bundle/sdk-minimal/README.md) instead lists one bundle that owns its complete explicit tree. Generic `dsh --profile <name>` hands its remaining arguments to that profile's command-line startup row: Web owns its flag family, headless owns its task positional, and the protocol profiles accept no app options. Patch overlays use launcher-owned `--patch`. A new, non-shipped target can use `--from-default-profile <template>` to copy one default template's bundle list and patch-reload policy before boot or config dump. This creates an independent profile with empty dependencies and an empty user patch: it neither reads a local profile named by the template nor records an inheritance relationship. The launcher claims the complete target directory exclusively, so existing state and concurrent creators fail without modification. `dsh plugin --profile <name> <args...>` is a thin pnpm forwarder that initializes a base-backed profile and reconciles `dsh.profile.bundles` with installed bundle declarations; a package without a bundle declaration remains a plain dependency. [Headless as a direct core entry point](../../archived/architecture/2026-08-09-headless-direct-core-entry-point.md) owns the headless composition contract.
Resolution is two-anchored by construction: `dsh.profile.bundles` names resolve from the dsh installation first, then the profile directory — so in-box bundles always come from the same installation as the running `dsh` and pnpm never manages them — while bare plugin names in patch rows resolve through the profile directory's Node parent-walk into the maintained flat fallback `$DSH_HOME/profiles/node_modules` (one symlink per package the installation's app and bundles depend on, healed on every launch).
Resolution is two-anchored by construction: `dsh.profile.bundles` names resolve from the dsh installation first, then the profile directory, so in-box bundles always come from the same installation as the running `dsh` and pnpm never manages them. Bare plugin names in patch rows use the [immutable profile resolution generation](2026-09-09-profile-resolution-generations.md), which applies the same installation-first and ordered-bundle rules in memory; retained link and dual modes can materialize the same result.
Two supporting refactors: the webserver's built-in static dist serving became the single-owner **fallback seat** (`registerFallback`/`applyIndexTaps`), with the SPA server extracted to `@deepseek-ai/dsh-host-frontend-static` so the web bundle owns its dist as composition, not launcher code; and the personal-overlay machinery of the [dsh CLI personal-config decision](../../archived/feature/2026-07-20-dsh-cli-personal-config.md) (`loadPersonalPatches`, `$DSH_HOME/config.yaml`) was retargeted to the per-profile and home-level `cordis.patch.yml` layers (`loadOptionalPatches`, `watchUserPatches` taking a filename), superseding that note's entry modes and file location while keeping its Harness-home root, patch semantics, and fail-loud parsing.
@@ -31,5 +31,5 @@ Two supporting refactors: the webserver's built-in static dist serving became th
- New composition surfaces (a TUI, provider packs) ship as ordinary npm packages installable per profile, without a repository row for every deployment shape.
- Users can start an independent custom profile from any shipped application template without copying machine-local profile state.
- `apps/cli` shrank to argv parsing, profile machinery consumption, and the pnpm forwarder; `AppCLIEntry` and the per-surface boot paths are gone.
- The keyless web e2e scaffold boots the same bundle layers over the same empty-root shape as production, including the profiles module fallback, so composition drift between test and product fails loudly.
- The keyless web e2e scaffold boots the same bundle layers over the same empty-root shape as production and exercises the same profile package-selection rules, so composition drift between test and product fails loudly.
- Under the pre-release stance, backends carry no compatibility behavior for old on-disk configuration; `$DSH_HOME/config.yaml` is ignored.
@@ -14,7 +14,7 @@ Status: implemented
默认 Profile 模板为 `web`、`headless`、`sdk` 与 `acp` 使用 `@deepseek-ai/dsh-base` 作为共享核心,并在其上叠加一个模式组合包。[独立 `sdk-minimal` profile](../../../../packages/bundle/sdk-minimal/README.zh.md)则只列出一个拥有完整显式配置树的组合包。通用的 `dsh --profile <name>` 把剩余参数交给该 profile 的命令行启动行:Web 持有自己的 flag 家族,headless 持有任务位置参数,协议 profile 不接受应用选项。patch overlay 使用启动器持有的 `--patch`。新的非内置目标可以使用 `--from-default-profile <template>`,在启动或配置 dump 之前复制一个默认模板的 bundle 列表与 patch 重载策略。这会创建依赖为空、用户 patch 为空的独立 profile:它既不读取与模板同名的本地 profile,也不记录继承关系。launcher 会以独占方式领取完整的目标目录,因此既有状态和并发创建者都会在不作修改的情况下失败。`dsh plugin --profile <name> <args...>` 是一层薄薄的 pnpm 转发器,负责初始化一个以 base 为基础的 profile,并依据已安装包的组合包声明调和 `dsh.profile.bundles`;没有组合包声明的包保持为普通依赖。[Headless 作为直接 core 入口](../../archived/architecture/2026-08-09-headless-direct-core-entry-point.md)负责 headless 组合约定。
解析在构造上就是双锚点的:`dsh.profile.bundles` 中的名称先从 dsh 安装目录解析,再从 profile 目录解析——因此内置组合包始终来自与运行中 `dsh` 相同的安装,pnpm 从不管理它们——而 patch 行中的裸插件名称经 profile 目录的 Node 父目录逐级查找,落到受维护的扁平回退目录 `$DSH_HOME/profiles/node_modules`(安装目录的应用与各组合包所依赖的每个包各一个符号链接,每次启动时修复)。
解析在构造上就是双锚点的:`dsh.profile.bundles` 中的名称先从 dsh 安装目录解析,再从 profile 目录解析,因此内置组合包始终来自与运行中 `dsh` 相同的安装,pnpm 从不管理它们。patch 行中的裸插件名称使用[不可变 profile resolution generation](2026-09-09-profile-resolution-generations.zh.md),在内存中应用相同的安装优先与有序 bundle 规则;保留的 link 与 dual 模式可以物化同一结果。
两项配套重构:webserver 内置的静态 dist 服务改为单一所有者的**回退席位**(`registerFallback`/`applyIndexTaps`),SPA 服务器提取到 `@deepseek-ai/dsh-host-frontend-static`,使 web 组合包以组合的方式持有自己的 dist,而不是靠启动器代码;[dsh CLI 个人配置决策](../../archived/feature/2026-07-20-dsh-cli-personal-config.md)的个人 overlay 机制(`loadPersonalPatches`、`$DSH_HOME/config.yaml`)改为面向逐 profile 与 home 级的 `cordis.patch.yml` 层(`loadOptionalPatches`、接受文件名的 `watchUserPatches`),取代该笔记的各入口模式与文件位置,同时保留其 Harness home 根目录、patch 语义与响亮失败的解析。
@@ -31,5 +31,5 @@ Status: implemented
- 新的组合表层(TUI、提供方扩展包)以普通 npm 包形式交付,可按 profile 安装,无需在仓库中为每种部署形态各留一行。
- 用户可以从任意随附应用模板启动一个独立的自定义 profile,而不会复制机器本地的 profile 状态。
- `apps/cli` 收缩为 argv 解析、profile 机制的消费方和 pnpm 转发器;`AppCLIEntry` 与各表层专属的启动路径全部移除。
- 无密钥 web e2e 脚手架以与生产相同的空根形态启动相同的组合包层,包括 profiles 模块回退,因此测试与产品之间的组合漂移会响亮失败。
- 无密钥 web e2e 脚手架以与生产相同的空根形态启动相同的组合包层,并执行相同的 profile 包选择规则,因此测试与产品之间的组合漂移会响亮失败。
- 按发布前姿态,后端不携带旧磁盘配置的兼容行为;`$DSH_HOME/config.yaml` 会被忽略。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-09-client-conversation-node-assembly.md
2026-08-09-client-conversation-node-assembly.md: 842dfb0218478591f975c97064f101a35ea2f211
2026-08-09-client-conversation-node-assembly.zh.md: 6ab64c25957d33487461fe56e122faedb19f9421
2026-08-09-client-conversation-node-assembly.md: 43f9f0359822f7d7c5b52c09646fd3ffbf80f783
2026-08-09-client-conversation-node-assembly.zh.md: d7475f5ff8218419743448d455d5d5ffe3cc4091
@@ -263,7 +263,7 @@ Page size, record packing, the number of history loads, and RAF coalescing affec
| Message / `input-message` | Message ID | Append-surface `user/message` | None | Use source for a context message, or read the nearest next-step Inbox to distinguish user from steering |
| Request Prompt / `request-prompt` | Header Event seq | Each `request/header` | None | Read the preceding Request Prompt through Reader, retain the full prompt state, and classify system/tool changes |
| Assistant / `assistant-step` | `turn:step` | `step/start` | Live `assistant/live-chunk`, durable `assistant/message` or `assistant/attempt`, and same-step Retry | Aggregate blocks, usage, first-token time, settlement evidence, and retry-hidden state, then publish same-key Step data |
| Tool / `tool-call` | Root call ID | Root `tool/call` | Root result and Code Dispatch start/result | Aggregate the root, children, and parent Map; Dispatch Events route exactly through `rootCallId` |
| Tool / `tool-call` | Root call ID | Root `tool/call` | Root result and PTC dispatch start/result | Aggregate the root, children, and parent Map; Dispatch Events route exactly through `rootCallId` |
| Command / `command` | Command ID | `command/run` | `command/done` and compact lifecycle/checkpoint Events carrying a source command ID | Aggregate command outcome and manual-compaction evidence |
| Automatic Compaction / `compaction` | Compaction ID | `compaction/start` without a source command ID | Summary, end, and replacement checkpoint | Aggregate summary/checkpoint; sufficient checkpoint evidence supports fallback without a start |
| Retry / `model-retry` | Retry ID | Attempt 1 `llm/retry` | Later `llm/retry` and `llm/retry-started` | Aggregate one RetryId's attempts and scheduled/started state |
@@ -360,7 +360,7 @@ SessionEventLike window
Runtime tests pin Definition lifecycle registration, exact-ID append, update-before-start collection followed by forward replay after start, prepend identity, Reader window-gap repair, transitive dependencies, Location closure, Step→Turn data phase order, Location data replacement, publication cadence, illegal withdrawal, first-subscription activation, monotonic active targets, and per-target Builders.
Conversation tests cover every built-in Chat Definition, Assistant Step data, Turn Tail and Deliverables Turn data, Chat ordering and structural sharing, selector isolation, Assistant and Tool running-to-settled identity, nested Code Dispatch, steering, Compaction, Retry, interruption, load-older anchoring, and slot dispatch. Trajectory tests cover its independently registered Message, Assistant, Tool, Compaction, Request-header, and boundary Definitions together with the preserved stage-oriented view model.
Conversation tests cover every built-in Chat Definition, Assistant Step data, Turn Tail and Deliverables Turn data, Chat ordering and structural sharing, selector isolation, Assistant and Tool running-to-settled identity, nested PTC dispatch, steering, Compaction, Retry, interruption, load-older anchoring, and slot dispatch. Trajectory tests cover its independently registered Message, Assistant, Tool, Compaction, Request-header, and boundary Definitions together with the preserved stage-oriented view model.
Slot type/runtime tests pin required parent-provided common inject, the `hookContext` type, Hook isolation across Node contexts, stable factory/Hook identity, and the absence of business-renderer rerenders for unrelated Session publications. Existing entry-owned Observable Hook tests continue to pin the path that does not use a contextual factory.
@@ -263,7 +263,7 @@ Chat `order` 的结构性变化仍可能重排当前可见 key;纯 data 更新
| Message / `input-message` | message ID | append-surface `user/message` | 无 | 根据 source 生成 context message,或读取最近 next-step Inbox 判断 user/steering |
| Request Prompt / `request-prompt` | header Event seq | 每条 `request/header` | 无 | 通过 Reader 读取前一条 Request Prompt,保留完整 prompt 状态,并判定 system/tool 变化 |
| Assistant / `assistant-step` | `turn:step` | `step/start` | Live `assistant/live-chunk`、持久 `assistant/message` 或 `assistant/attempt`、同 step Retry | 聚合 block、usage、首 token 时间、settlement 证据与 retry-hidden state,再发布同 key Step data |
| Tool / `tool-call` | root call ID | root `tool/call` | root result、Code Dispatch start/result | 聚合 root、children 和 parent Map;Dispatch Event 用 `rootCallId` 精确路由 |
| Tool / `tool-call` | root call ID | root `tool/call` | root result、PTC dispatch start/result | 聚合 root、children 和 parent Map;Dispatch Event 用 `rootCallId` 精确路由 |
| Command / `command` | command ID | `command/run` | `command/done`、带 source command ID 的 compact lifecycle/checkpoint | 聚合 command outcome 和手动压缩证据 |
| Automatic Compaction / `compaction` | compaction ID | 无 source command ID 的 `compaction/start` | summary、end、replacement checkpoint | 聚合 summary/checkpoint;checkpoint 足够时可在缺 start 下 fallback |
| Retry / `model-retry` | retry ID | attempt 1 的 `llm/retry` | 后续 `llm/retry` 与 `llm/retry-started` | 聚合同一 RetryId 的 attempts 与 scheduled/started 状态 |
@@ -360,7 +360,7 @@ SessionEventLike window
Runtime tests 固定 Definition 生命周期注册、exact-ID append、update-before-start 收集与 start 后正序 replay、prepend identity、Reader window-gap 修复、传递依赖、Location closure、Step→Turn data phase order、Location data replacement、publication cadence、非法撤回、首次订阅 activation、单调 active target 和 per-target Builder。
Conversation tests 覆盖全部内建 Chat Definition、Assistant Step data、Turn Tail 与 Deliverables Turn data、Chat 排序和结构共享、selector isolation、Assistant/Tool running-to-settled identity、nested Code Dispatch、steering、Compaction、Retry、interruption、load-older anchoring 和 slot dispatch。Trajectory tests 则覆盖它独立注册的 Message、Assistant、Tool、Compaction、Request-header 与 boundary Definition,以及继续保留的 stage-oriented view model。
Conversation tests 覆盖全部内建 Chat Definition、Assistant Step data、Turn Tail 与 Deliverables Turn data、Chat 排序和结构共享、selector isolation、Assistant/Tool running-to-settled identity、nested PTC dispatch、steering、Compaction、Retry、interruption、load-older anchoring 和 slot dispatch。Trajectory tests 则覆盖它独立注册的 Message、Assistant、Tool、Compaction、Request-header 与 boundary Definition,以及继续保留的 stage-oriented view model。
Slot type/runtime tests 固定父注册必须提供声明的 common inject、`hookContext` 类型、不同 Node context 的 Hook 隔离、factory/Hook identity 稳定,以及无关 Session publication 不重渲染业务 renderer。原 entry-owned Observable Hook 测试继续固定未使用 contextual factory 的路径。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-10-session-log-version-mechanism.md
2026-08-10-session-log-version-mechanism.md: 0f7f70b5ad6ecb2445729b1aa61fb3b295fc4ddb
2026-08-10-session-log-version-mechanism.zh.md: a6d58505ad9fdb6068a1afe48f7250f020d20a9e
2026-08-10-session-log-version-mechanism.md: 513b126d3b00841f71893715cc296cb67619543e
2026-08-10-session-log-version-mechanism.zh.md: 919bfb077c7037da2ebf7eb81b75745d6b4fe4d0
@@ -16,7 +16,13 @@ Session logs must be upgradable after release, and the runtime that ships first
**Read rules by direction.** Equal version: read normally. Newer than the reader: refuse, name the direction ("written by a newer harness — upgrade"), and point at the raw log artifact so the user can still see the text (`SessionFormatUnsupportedError`, distinct from `SessionPersistenceCorruptionError` because nothing is damaged). Older than the reader: every event-body operation first runs the complete adjacent chain in memory and leaves the source path, bytes, and inode unchanged. Read handles may consume that current logical result directly; a write open exclusively publishes the final current generation under its canonical versioned filename before append. Header-only listing remains non-mutating and reports the numerically highest canonical generation. Catalog generation and module initialization reject a missing adjacent step, so a published first-party build never exposes a partial historical chain. Retained lower generations are not automatic fallback or a downgrade compatibility promise.
**A per-event `ignorable` marker covers vocabulary growth, so ordinary event additions never bump the version.** The event vocabulary is decided by which plugins are mounted, which a single version integer cannot describe. A reader meeting an unrecognized event type refuses to interpret the log unless the event carries `ignorable: true` in its envelope. The default is *required*: forgetting the marker over-refuses a resumable session (an inconvenience), while a default of ignorable would make the same mistake silently resume a gutted one (a safety failure). The architecture makes this sound: model-visible content flows only through the three `surfaceOp`-marked surface event types plus the `request/header`/`request/context` folds, so the dangerous unknowns are exactly the non-surface events that change how the rest of the log is read (`session/end-seed` is the existing example).
**A per-event `ignorable` marker covers vocabulary growth, so ordinary event additions never bump the version.** The event vocabulary is decided by which plugins are mounted, which a single version integer cannot describe. A reader meeting an unrecognized event type refuses to interpret the log unless the event carries `ignorable: true` in its envelope. The default is *required*: forgetting the marker over-refuses a resumable session (an inconvenience), while a default of ignorable would make the same mistake silently resume a gutted one (a safety failure). The architecture makes this sound: model-visible content flows only through the four `surfaceOp`-marked surface event types plus the `request/header`/`request/context` folds, so the dangerous unknowns are exactly the non-surface events that change how the rest of the log is read (`session/end-seed` is the existing example).
### Writer and publication authority
`SESSION_FORMAT_VERSION` owns the checkout writer number; the [release-status reference](../../../../docs/session-format-status.md) owns one bilingual `latestReleasedVersion` and `evidenceTag` record. Publication changes independently of source development, so status is derived by comparing those facts rather than maintaining a second `released` boolean. General documentation links to these authorities; fixed-version contracts and historical evidence keep their explicit numbers.
The [documentation-standard check](../../../../scripts/doc-standard.spec.ts) validates record structure, bilingual equality, evidence-link consistency, and the local release/writer ordering without network access. It proves internal consistency, not publication or freshness. The release operator verifies publication and updates the record after a higher format ships, as required by the [release process](../process/2026-08-10-npm-release-sequences.md). This keeps compatibility review independent of credentials and GitHub availability while making the manual freshness obligation explicit.
## Consequences
@@ -28,3 +34,5 @@ What shipped in v0 (release 0812): direction-aware refusal with the raw-log path
- **Default-ignorable unknown events** — inverts the failure mode of a forgotten marker from visible over-refusal into silent corruption.
- **Migrating during header-only listing** — makes cheap inventory mutate storage and requires event bodies to compute facts that a header cannot prove. Listing returns descriptors; event-body reads own publication.
- **Per-plugin runtime registration of known event types** — rejected because it would make the known set composition-dependent and register event names without classifying whether omission is safe. The persisted `ignorable` marker keeps that classification with each record; the [external-plugin retention decision](2026-08-30-retain-ignorable-external-session-events.md) owns the current consumer constraint.
- **Duplicate release flags or runtime status services** — introduce another mutable authority for a maintainer fact that does not control Session execution. The writer constant and publication record suffice.
- **Network-dependent documentation gates or publication automation** — network queries would couple local documentation checks to credentials and GitHub availability; a runtime service or publication workflow change is unnecessary for record consistency. Publication verification remains an explicit release-operator obligation.
@@ -16,7 +16,13 @@ Session log 在发布后必须能升级格式,而最先发布的运行时决
**读取规则按方向区分。**版本相等:正常读。比读取器新:拒绝,说明方向("由更新的 harness 写入,请升级"),并给出原始日志文件的路径,用户仍能看到文本(`SessionFormatUnsupportedError`,与 `SessionPersistenceCorruptionError` 区分,因为数据没有损坏)。比读取器旧:每个事件正文操作先在内存中运行完整相邻链,并保持源路径、字节与 inode 不变。读句柄可以直接使用该 current 逻辑结果;写 open 则在 append 前把最终 current generation 排他发布到其规范版本文件名。仅 header 的列表保持不变更,并报告数值最高的规范 generation。catalog 生成与模块初始化会拒绝缺失的相邻步骤,因此已发布第一方 build 绝不会暴露不完整历史链。保留的低 generation 不是自动 fallback,也不构成 downgrade compatibility 承诺。
**逐事件的 `ignorable` 标记吸收词汇表增长,普通的新增事件永远不用升版本。**事件词汇表由挂载了哪些插件决定,单个版本整数描述不了它。读取器遇到不认识的事件类型时拒绝解读日志,除非该事件的信封带 `ignorable: true`。默认为必需:忘写标记的后果是把一个本可恢复的会话拒绝过头(体验问题),而默认可忽略会让同样的疏忽静默恢复出残缺会话(安全事故)。架构保证了这条规则成立:模型可见内容只经三种带 `surfaceOp` 标记的 surface 事件加 `request/header`、`request/context` 折叠进入重建,危险的未知事件恰好是那些不进 surface 但改变日志其余部分解读方式的事件(`session/end-seed` 是现存例子)。
**逐事件的 `ignorable` 标记吸收词汇表增长,普通的新增事件永远不用升版本。**事件词汇表由挂载了哪些插件决定,单个版本整数描述不了它。读取器遇到不认识的事件类型时拒绝解读日志,除非该事件的信封带 `ignorable: true`。默认为必需:忘写标记的后果是把一个本可恢复的会话拒绝过头(体验问题),而默认可忽略会让同样的疏忽静默恢复出残缺会话(安全事故)。架构保证了这条规则成立:模型可见内容只经四种带 `surfaceOp` 标记的 surface 事件加 `request/header`、`request/context` 折叠进入重建,危险的未知事件恰好是那些不进 surface 但改变日志其余部分解读方式的事件(`session/end-seed` 是现存例子)。
### 写入器与发布真源
`SESSION_FORMAT_VERSION` 拥有工作区写入器版本号;[发布状态参考](../../../../docs/session-format-status.zh.md)拥有唯一的双语 `latestReleasedVersion` 与 `evidenceTag` 记录。发布状态独立于源码开发而变化,因此通过比较这两个事实推导状态,而不另行维护 `released` 布尔值。一般文档链接到这些真源;固定版本约定与历史证据保留明确版本号。
[文档标准检查](../../../../scripts/doc-standard.spec.ts)在不访问网络的情况下,校验记录结构、双语一致性、证据链接一致性及本地发布版本与写入器版本的大小关系。它证明内部一致性,而非发布事实或记录新鲜度。[发布流程](../process/2026-08-10-npm-release-sequences.zh.md)要求发布操作者在更高格式交付后核实发布并更新记录。这让兼容性评审不依赖凭据与 GitHub 可用性,同时明确人工维护新鲜度的义务。
## 影响
@@ -28,3 +34,5 @@ v0(0812 发布)交付的内容:分方向的拒绝并带原始日志路径
- **未知事件默认可忽略**:把忘写标记的后果从可见的过度拒绝反转成静默损坏。
- **在仅 header 列表期间迁移**:让便宜清单改变存储,而且需要读取事件正文才能计算 header 无法证明的事实。列表返回 descriptor,事件正文读取负责发布。
- **插件运行时注册已知事件类型**:不予采用,因为该方案会让已知集依赖插件组合,而且只注册事件名称,无法判定省略事件是否安全。持久化的 `ignorable` 标记把该分类保留在每条记录中;[外部插件保留决策](2026-08-30-retain-ignorable-external-session-events.zh.md)定义当前消费方约束。
- **重复发布标记或运行时状态服务**:为不控制 Session 执行的维护信息增加另一个可变真源。写入器常量与发布记录已经足够。
- **依赖网络的文档门禁或发布自动化**:网络查询会把本地文档检查耦合到凭据与 GitHub 可用性;记录一致性不需要运行时服务或发布工作流变更。核实发布仍是发布操作者的明确义务。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-15-client-shells-and-dynamic-packages.md
2026-08-15-client-shells-and-dynamic-packages.md: 1d67c778b6a06849324dd6095a98d57dc41b94f9
2026-08-15-client-shells-and-dynamic-packages.zh.md: db1e4e7e7b319c283ae39a94535d88d4dc71d60a
2026-08-15-client-shells-and-dynamic-packages.md: 89f89784512b86b70ee1b9460850e6f2f951a082
2026-08-15-client-shells-and-dynamic-packages.zh.md: 5512d9262e979a94a65c25a6c1271b030e23ec70
@@ -51,7 +51,7 @@ The modules Node half injects the startup protocol into the served HTML in this
4. Assign `window.__DSH_BOOT__`, including all scheduling descriptors and every row's one-resource HMR combo URL.
5. Execute the Vite main module.
The bootstrap combo currently registers only the modules factory. The startup kernel passes the raw graph and shell seeds to `__ModuleLoader__.create()`. The facade removes the modules registration, materializes it with a `require` function that rejects every external, and invokes its `createClientModuleSystem` export. The modules bundle parses the graph, constructs `ClientModuleSystem`, caches its own exports as the modules row, retains the system in a module closure, and switches the same facade to live mode. The modules client face consequently has a zero-external bootstrap requirement.
The bootstrap combo currently registers only the modules factory. The startup kernel passes the raw graph and shell seeds to `__ModuleLoader__.create()`. The facade removes the modules registration, materializes it with a `require` function that rejects every external, and invokes its `createClientModuleSystem` export. The modules bundle parses the graph, constructs and returns `ClientModuleSystem`, caches its own exports as the modules row, and switches the same facade to live mode. The kernel installs that instance as its Loader's `internal`, and the modules plugin reads it there when it provides `ctx.modules`. The modules client face consequently has a zero-external bootstrap requirement and no module-global system identity.
After the `immediately` tier has registered its factories, the kernel creates all Loader entries, awaits Cordis quiescence, and requires every fiber to be ACTIVE. It then calls `ctx.uiRenderer.mount(container)`. The dynamic `ui-renderer` package owns React, slot rendering, hydration of the existing boot DOM, and the React root lifecycle; the startup kernel and failure page remain React-free.
@@ -51,7 +51,7 @@ Modules Node 半按以下顺序向实际返回的 HTML 注入启动协议:
4. 赋值 `window.__DSH_BOOT__`,其中包含全部调度描述及每个 row 的单资源 HMR combo URL。
5. 执行 Vite 主模块。
Bootstrap combo 当前只登记 modules factory。启动内核把原始图与外壳 seed 传给 `__ModuleLoader__.create()`。Facade 移除 modules registration,用拒绝全部 external 的 `require` 函数将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造 `ClientModuleSystem`、把自身 exports 缓存为 modules row、在模块闭包中保留该系统,并把同一 facade 切换到 live 模式。因此 modules client face 必须满足零 external 的自举要求。
Bootstrap combo 当前只登记 modules factory。启动内核把原始图与外壳 seed 传给 `__ModuleLoader__.create()`。Facade 移除 modules registration,用拒绝全部 external 的 `require` 函数将其物化,再调用其 `createClientModuleSystem` 导出。Modules bundle 解析图、构造并返回 `ClientModuleSystem`、把自身 exports 缓存为 modules row,并把同一 facade 切换到 live 模式。内核把该实例装成自身 Loader 的 `internal`,modules 插件从这里读取并提供 `ctx.modules`。因此 modules client face 保持零 external 的自举要求,也没有模块级系统身份。
`immediately` 层级完成 factory 注册后,内核创建全部 Loader entry,等待 Cordis 静止,并要求每个 fiber 都进入 ACTIVE。随后调用 `ctx.uiRenderer.mount(container)`。动态 `ui-renderer` 包拥有 React、slot 渲染、已有启动 DOM 的 hydrate 和 React root 生命周期;启动内核与失败页保持 React-free。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-18-experimental-agent-teams-packages.md
2026-08-18-experimental-agent-teams-packages.md: bff91f9c735f541eb6ef055a5bf70c31b5d88912
2026-08-18-experimental-agent-teams-packages.zh.md: f3753a3be9679962c30b566a5a411ca568e1c982
2026-08-18-experimental-agent-teams-packages.md: 4a78a60c2ad7463c06b670e6ec664579ffd767b9
2026-08-18-experimental-agent-teams-packages.zh.md: 8056cd8194695e8ce41acb359e51c040d4ed3aec
@@ -1,4 +1,4 @@
# Agent Note: Incubate Agent Teams as private experimental packages
# Agent Note: Publish Agent Teams under experimental package names
Status: implemented
@@ -6,34 +6,36 @@ English | [中文](2026-08-18-experimental-agent-teams-packages.zh.md)
## Problem
Agent Teams needs the real Session log, subagent lifecycle, tools, examples, snapshots, and repository checks while its service and tool contracts continue to change. Placing those packages in a product-role group makes them members of the dsh release family and gives them the same publication expectation as stable packages.
Agent Teams needs the real Session log, subagent lifecycle, tools, examples, snapshots, and repository checks while its service and tool contracts continue to change. Users also need to install the complete Team composition from npm without building a source checkout.
An experimental directory without a current package previously imposed placement, dependency, promotion, and release rules on no consumer. Agent Teams supplies the concrete consumer, but the directory needs mechanical release exclusion and dependency isolation rather than a documentation-only status.
Moving the packages into product-role groups would remove their experimental names and imply stable-package ownership. Publishing every package under `packages/experimental/` would instead expose unrelated internal prototypes. The release policy must let users install Agent Teams while keeping internal-only prototypes private.
## Decision
`packages/experimental/agent-team`, `packages/experimental/tool-agent-team`, `packages/experimental/agent-team-profile`, `packages/experimental/client-ui-agent-team`, and `packages/experimental/agent-team-web-profile` are private workspace packages. The [experimental package rules](../../../../packages/experimental/AGENTS.md) own their npm names and promotion rename; this note owns their placement, release exclusion, and dependency isolation.
`packages/experimental/agent-team`, `packages/experimental/tool-agent-team`, `packages/experimental/agent-team-profile`, `packages/experimental/client-ui-agent-team`, and `packages/experimental/agent-team-web-profile` are public workspace packages. They retain their existing `@deepseek-ai/dsh-experimental-*` names and join the dsh release family. The [publication denylist decision](../process/2026-09-12-experimental-publication-denylist.md) owns the public default and private exceptions; the [experimental package rules](../../../../packages/experimental/AGENTS.md) own dependency isolation and later promotion.
The dsh pack and publish set and the local baseline publisher exclude every manifest below `packages/experimental/`. `release:dsh` still advances their manifest versions with the shared dsh version without creating release tags. Workspace constraints require each experimental package to set `private: true` and omit `publishConfig`. The same top-level check rejects `dependencies`, `optionalDependencies`, and `peerDependencies` from release packages, release apps, or the Python runtime to an experimental package. Experimental packages may depend on release packages and each other; tests may use them through `devDependencies`, and examples may load them explicitly.
The dsh pack and publish set and the local baseline publisher include these five Agent Teams directories and the [Cua Driver providers](2026-09-12-computer-use-provider-registration.md). Workspace constraints require them to omit `private`, set `publishConfig.access` to `public`, and keep the experimental npm prefix. Release packages and apps outside the experimental group, plus the Python runtime, cannot name experimental packages in `dependencies`, `optionalDependencies`, or `peerDependencies`; experimental packages may depend on release packages and each other.
The generic caller-reserved continuable child identity and selective direct-child drain remain in the stable Subagent service. They own Subagent identity and Activation lifecycle without importing or naming Agent Teams; the experimental Team service consumes them in the permitted direction.
The private Host-side Agent Teams profile bundle depends on the Team packages and applies after `dsh-base`. It inserts the Team rows and disables the global continuable-child controls whose model-visible names overlap the Team tools. The separate private Web profile applies after `dsh-web-app` and the Host profile; it inserts the Team UI, which mounts the Remote contribution generated by the Team package. Both layers leave the shipped base, CLI, Web, and Python runtime dependency graphs unchanged.
The published Host-side Agent Teams profile bundle depends on the Team packages and applies after `dsh-base`. It inserts the Team rows and disables the global continuable-child controls whose model-visible names overlap the Team tools. The separate published Web profile applies after `dsh-web-app` and the Host profile; it inserts the Team UI, which mounts the Remote contribution generated by the Team package. Both layers remain opt-in and leave the shipped base, CLI, Web, and Python runtime dependency graphs unchanged.
Profile startup resolves selected bundles before healing module fallbacks. The shared fallback retains the dsh installation's carrier-specific entries: symlinks under plain Node and ESM proxies in a packaged executable. Missing packages from selected bundle closures are linked under the current profile's own `node_modules`, while pnpm-managed profile entries remain authoritative. Closure discovery starts from each explicit external bundle's real package directory and traverses every listed root even when an earlier dependency has the same package name. It excludes dsh-owned profile projections from later discovery, so a projected dependency cannot feed back into its own closure. Link ownership compares canonical parent paths so junction-normalized targets remain removable. A private profile layer can therefore carry experimental plugin rows without adding those plugins to a release app, requiring profile users to install transitive packages directly, weakening packaged-runtime module identity, or changing another profile's resolution.
Profile startup resolves selected bundles before computing the [immutable profile resolution generation](2026-09-09-profile-resolution-generations.md). The generation retains installation-first precedence, traverses each explicit bundle root completely in profile order, and keeps pnpm-managed profile packages authoritative. Runtime mode enforces the result in memory; retained link and dual modes materialize the same result as shared and profile-owned projections. A private profile layer can therefore carry experimental plugin rows without adding those plugins to a release app, requiring profile users to install transitive packages directly, weakening packaged-runtime module identity, or changing another profile's resolution.
Experimental status changes publication and compatibility expectations only. The packages retain the repository's ordinary documentation, invariant, lifecycle, security, unit, real-composition, and snapshot requirements. Promotion requires review of the public contracts, limitations, test evidence, release payload, runtime dependents, and a named owner accepting stable-package obligations.
Experimental status changes compatibility and support expectations, not publication for these five packages. They retain the repository's ordinary documentation, invariant, lifecycle, security, unit, real-composition, and snapshot requirements. Promotion still requires review of the public contracts, limitations, test evidence, runtime dependents, and a named owner accepting stable-package obligations.
## Alternatives considered
**Keep Agent Teams in a product-role group and describe it as opt-in.** Opt-in composition controls model behavior but does not exclude packages from publication or prevent stable packages from taking runtime dependencies on them.
**Move Agent Teams into product-role groups.** This would remove the requested experimental npm names and imply stable-package ownership before the contracts have stabilized.
**Reserve an empty experimental group.** A directory without a current package has no owner or release mechanism to test. The group exists only while concrete packages need its enforced treatment.
**Keep Agent Teams private and source-checkout only.** This preserves the simplest experimental policy but prevents users from installing the complete opt-in composition from npm.
**Publish every experimental package.** Unrelated prototypes remain internal-only and have not accepted a public package contract.
**Move the Subagent prerequisites into the experimental directory.** Child identity allocation and Activation teardown belong to the Subagent owner and contain no Team-specific contract. Moving or duplicating them would invert the dependency or split one lifecycle across packages.
## Consequences
Agent Teams can use the full repository graph and quality checks without entering official tarballs or becoming a supported runtime dependency. A release package cannot expose Team until the Team packages are promoted, so the CLI experiment installs an explicit private profile layer instead of changing shipped bundles. The generic profile launcher accepts that layer without making its plugin dependencies part of the dsh release closure.
Agent Teams publishes as five installable tarballs in the dsh release family without changing package names or enabling Team in a shipped profile. Public availability does not make the packages stable or supported by default, and stable release packages cannot take runtime dependencies on them.
The product-role grouping is less direct while the packages incubate. Promotion creates path and npm-name churn as specified by the experimental package rules.
The release family carries the experimental npm names. Promotion still creates path and npm-name churn as specified by the experimental package rules.
@@ -1,4 +1,4 @@
# Agent Note: 将 Agent Teams 作为私有实验性包孵化
# Agent Note: 以实验性包名发布 Agent Teams
Status: implemented
@@ -6,34 +6,36 @@ Status: implemented
## 问题
Agent Teams 的服务与工具约定仍在变化,但它需要使用真实 Session 日志、subagent 生命周期、工具、示例、快照和仓库检查。把这些包放在产品职责组会使其成为 dsh 发布系列成员,并获得与稳定包相同的发布预期。
Agent Teams 的服务与工具约定仍在变化,但它需要使用真实 Session 日志、subagent 生命周期、工具、示例、快照和仓库检查。用户还需要直接从 npm 安装完整 Team 组合,而无需构建源码 checkout。
没有实际包的 experimental 目录曾经让没有消费方的放置、依赖、promotion 和发布规则长期存在。Agent Teams 提供了具体消费方,但该目录需要机械强制的发布排除与依赖隔离,不能只用文档标记状态。
把这些包移入产品职责组会移除实验性名称,并暗示稳定包 owner 已经就位。发布 `packages/experimental/` 下的所有包又会暴露无关的内部原型。发布策略必须让用户安装 Agent Teams,同时让内部专用原型保持私有。
## 决策
`packages/experimental/agent-team`、`packages/experimental/tool-agent-team`、`packages/experimental/agent-team-profile`、`packages/experimental/client-ui-agent-team` 与 `packages/experimental/agent-team-web-profile` 是私有 workspace 包。[实验性包规则](../../../../packages/experimental/AGENTS.md)负责其 npm 名和 promotion 重命名;本记录负责其目录归属、发布排除与依赖隔离。
`packages/experimental/agent-team`、`packages/experimental/tool-agent-team`、`packages/experimental/agent-team-profile`、`packages/experimental/client-ui-agent-team` 与 `packages/experimental/agent-team-web-profile` 是公开 workspace 包。它们保留现有 `@deepseek-ai/dsh-experimental-*` 名称并加入 dsh 发布系列。[发布拒绝列表决策](../process/2026-09-12-experimental-publication-denylist.zh.md)负责默认公开与私有例外;[实验性包规则](../../../../packages/experimental/AGENTS.md)负责依赖隔离与后续 promotion。
dsh pack 与 publish 集合以及本地 baseline 发布器均排除 `packages/experimental/` 下的所有 manifest。`release:dsh` 仍会让这些 manifest 跟随 dsh 共享版本递增,但不会创建发布 tag。workspace 约束要求每个实验性包设置 `private: true` 并省略 `publishConfig`。同一个顶层检查会拒绝发布包、发布 app 或 Python runtime 通过 `dependencies`、`optionalDependencies` 或 `peerDependencies` 依赖实验性包。实验性包可以依赖发布包和其他实验性包;测试可以通过 `devDependencies` 使用它们,示例可以显式加载它们。
dsh 打包与发布集合以及本地基线发布器包含这五个 Agent Teams 目录和 [Cua Driver 提供方](2026-09-12-computer-use-provider-registration.zh.md)。workspace 约束要求它们省略 `private`、将 `publishConfig.access` 设为 `public`,并保留实验性 npm 前缀。实验组之外的发布包、应用和 Python 运行时不能在 `dependencies`、`optionalDependencies` 或 `peerDependencies` 中引用实验包;实验包可以依赖发布包和彼此。
通用的调用方预留 continuable child 身份和精确 direct-child drain 仍属于稳定 Subagent 服务。它们负责 Subagent 身份与 Activation 生命周期,不 import 或命名 Agent Teams;实验性 Team 服务沿允许的方向消费这些能力。
私有 Host 侧 Agent Teams profile bundle 依赖 Team 包,并在 `dsh-base` 之后应用。它会插入 Team 配置行,并禁用模型可见名称与 Team 工具重叠的全局 continuable-child control。独立的私有 Web profile 在 `dsh-web-app` 与 Host profile 之后应用;它会插入 Team UI,后者挂载 Team package 生成的 Remote contribution。两个层都保持已发布 base、CLI、Web 与 Python runtime 的依赖图不变。
公开发布的 Host 侧 Agent Teams profile bundle 依赖 Team 包,并在 `dsh-base` 之后应用。它会插入 Team 配置行,并禁用模型可见名称与 Team 工具重叠的全局 continuable-child control。独立公开发布的 Web profile 在 `dsh-web-app` 与 Host profile 之后应用;它会插入 Team UI,后者挂载 Team package 生成的 Remote contribution。两个层都保持显式启用,不改变随附 base、CLI、Web 与 Python runtime 的依赖图。
profile 启动会先解析所选 bundle,再修复模块 fallback。共享 fallback 保留 dsh 安装的载体专用条目:普通 Node 下使用 symlink,打包 executable 中使用 ESM proxy。仅由所选 bundle 闭包携带的缺失包会链接到当前 profile 自己的 `node_modules` 下,而 pnpm 管理的 profile 条目仍具有优先权。闭包发现从每个显式外部 bundle 的真实包目录开始;即使前一个依赖具有相同包名,也会遍历所有列出的根。后续发现会排除 dsh 所有的 profile projection,避免投影后的依赖重新进入自己的闭包。link ownership 通过规范化父路径比较,使 junction 规范化后的 target 仍可删除。因此,私有 profile 层可以携带实验性 plugin 配置行,而无需把这些 plugin 加入发布 app、要求 profile 用户直接安装传递依赖、破坏 packaged-runtime 的模块身份,或改变其他 profile 的解析结果。
profile 启动会先解析所选 bundle,再计算[不可变 profile resolution generation](2026-09-09-profile-resolution-generations.zh.md)。generation 保留安装优先顺序,按 profile 顺序完整遍历每个显式 bundle 根,并让 pnpm 管理的 profile 包保持优先。runtime 模式在内存中强制该结果;保留的 link 与 dual 模式把同一结果物化为共享和 profile 自有投影。因此,私有 profile 层可以携带实验性 plugin 配置行,而无需把这些 plugin 加入发布 app、要求 profile 用户直接安装传递依赖、破坏 packaged-runtime 的模块身份,或改变其他 profile 的解析结果。
实验性状态只改变发布与兼容性预期。这些包仍须满足仓库的一般文档、不变式、生命周期、安全、单元测试、真实组合测试和快照要求。promotion 前必须评审公开约定、限制、测试证据、发布 payload、运行时依赖方,并由一名具名 owner 接受稳定包义务。
对这五个包而言,实验性状态改变兼容性与支持预期,而不阻止发布。这些包仍须满足仓库的一般文档、不变式、生命周期、安全、单元测试、真实组合测试和快照要求。promotion 前仍须评审公开约定、限制、测试证据、运行时依赖方,并由一名具名 owner 接受稳定包义务。
## 曾考虑的替代方案
**把 Agent Teams 留在产品职责组,并标为显式启用。** 显式启用的组合可以控制模型行为,但不会阻止包发布,也不能阻止稳定包对其建立运行时依赖。
**把 Agent Teams 移入产品职责组。** 这会移除要求保留的实验性 npm 名称,并在约定稳定前暗示已有稳定包 owner。
**预留空的 experimental 组。** 没有实际包的目录没有 owner,也没有可供测试的发布机制。只有具体包需要这套强制处理时,该组才存在。
**让 Agent Teams 保持私有且仅供源码 checkout 使用。** 这会保留最简单的实验性策略,但用户无法从 npm 安装完整 opt-in 组合。
**发布所有实验性包。** 其他原型仍只供内部使用,也没有接受公开包约定。
**把 Subagent 前置能力移入 experimental 目录。** child 身份分配与 Activation teardown 属于 Subagent owner,且不包含 Team 专用约定。移动或复制这些能力会反转依赖方向,或把同一个生命周期拆到多个包中。
## 后果
Agent Teams 可以使用完整仓库依赖图与质量检查,而不进入正式 tarball,也不会成为受支持的运行时依赖。在 Team 包 promotion 前,发布包不能暴露 Team,因此 CLI 实验会安装显式的私有 profile 层,而不是修改已发布 bundle。通用 profile launcher 可以接受该层,而不会让它的 plugin 依赖进入 dsh 发布闭包。
Agent Teams 会作为 dsh 发布系列中的五个可安装 tarball 发布,同时保持包名不变,也不会在随附 profile 中启用 Team。公开可用不代表这些包稳定或默认受支持,稳定发布包也不能对其建立运行时依赖。
孵化期间的产品职责分组不够直接。promotion 会按照实验性包规则产生路径和 npm 名改动。
发布系列保留实验性 npm 名称。promotion 仍会按照实验性包规则产生路径和 npm 名改动。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-18-session-history-and-event-transport.md
2026-08-18-session-history-and-event-transport.md: 9a593ca9bfe4a80657d3eeeaa89cf0dcc73dd829
2026-08-18-session-history-and-event-transport.zh.md: a7f8b8f27153a1814ca65bbf9042d78717b50849
2026-08-18-session-history-and-event-transport.md: 8781ea265798ff200a6e0a58542b7d03693d8bd3
2026-08-18-session-history-and-event-transport.zh.md: 1603f5aed8adb7a42fecab92511176d2147be5c3
@@ -220,7 +220,7 @@ Session-list `updatedAt` is `max(header.createdAt, sessionListMetadata.lastPromp
`packages/api/workspace-controller` provides Host `ctx.workspaceController` and the generated `ctx.remote.workspace` namespace.
It owns create, rename, delete, insertBefore, insertSessionBefore, archiveSession, and `follow`. Workspace registry remains the durable source of truth; the Controller owns Remote commands, projection, and error mapping.
It owns create, rename, delete, insertBefore, insertSessionBefore, archiveSession, unarchiveSession, and `follow`. Workspace registry remains the durable source of truth; the Controller owns Remote commands, projection, and error mapping.
`WorkspaceFeed` synchronously observes storage `domain/changed`, and each follow generation emits a complete baseline before `upsert`, `remove`, `order`, and `archived` deltas.
@@ -220,7 +220,7 @@ Session 列表的 `updatedAt` 取 `max(header.createdAt, sessionListMetadata.las
`packages/api/workspace-controller` 提供 Host `ctx.workspaceController` 与生成的 `ctx.remote.workspace` namespace。
它拥有 create、rename、delete、insertBefore、insertSessionBefore、archiveSession 与 `follow`。Workspace registry 仍是持久事实来源,Controller 负责 Remote 命令、投影和错误映射。
它拥有 create、rename、delete、insertBefore、insertSessionBefore、archiveSession、unarchiveSession 与 `follow`。Workspace registry 仍是持久事实来源,Controller 负责 Remote 命令、投影和错误映射。
`WorkspaceFeed` 同步观察 storage `domain/changed`,并为每个 follow generation 先发送完整 baseline,再发送 `upsert`、`remove`、`order` 与 `archived` 增量。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-19-session-projection-mandatory-seam.md
2026-08-19-session-projection-mandatory-seam.md: 371faa254809649e93685a8f3ada3e640d5f391e
2026-08-19-session-projection-mandatory-seam.md: e1c6835311228fcaf7a22140f1db8da748603186
2026-08-19-session-projection-mandatory-seam.zh.md: 5d6a5169a556febae3f5c960b2d519c324d99863
@@ -25,7 +25,7 @@ The registry provides `stateOf(session, key)` for one typed host state and keeps
- **Default missing projected state.** This preserves more partial compositions but makes missing host state indistinguishable from a valid empty value. Rejected because official profiles mount the registry and configuration errors must fail explicitly.
- **Require every contributor at activation.** This makes the key set uniform but unnecessarily couples contribution lifecycle to service activation. Explicit first-access failure preserves the optional registration form without allowing silent degradation.
- **Use `snapshot()` for every read.** This keeps one method but computes unrelated wire views and encourages consumers to depend on batch transport data for host logic. Rejected in favor of typed single-key state reads.
- **Send full host values to clients.** This avoids separate view types but exposes provenance and policy knobs that no client consumes. Rejected in favor of explicit cropped views.
- **Send full host values to clients.** This avoids separate view types but exposes producer identities and policy knobs that no client consumes. Rejected in favor of explicit cropped views.
- **Broadcast registry additions and removals across Host and mux streams.** The streams have no shared ordering, so clients need tombstones, buffered frames, and baseline retries to reconcile them. Rejected because plugin-key churn does not justify a second synchronization protocol.
## Consequences
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-21-deepseek-llm-api-request-extensions.md
2026-08-21-deepseek-llm-api-request-extensions.md: eadbe2a5f17de446c345120f9a6aeeeb531c43b4
2026-08-21-deepseek-llm-api-request-extensions.zh.md: f45210adbc075c30484754ce7f8c2b65b515be6c
2026-08-21-deepseek-llm-api-request-extensions.md: 7a3609a3f7d8b2aa6edb634643ef234b7e3e8c07
2026-08-21-deepseek-llm-api-request-extensions.zh.md: 69426df3cc85851e448b98dc51af79555b8cbafc
@@ -14,13 +14,13 @@ Both values belong only on the official DeepSeek adapter path. Adding them to `G
## Decision
`@deepseek-ai/dsh-deepseek-llm-api-extensions` registers `ctx.deepseekLlmApiExtensions`, an additive registry of top-level fields for `deepseek-official` request bodies. A contributor claims one declaration-merged field with `register()`. The adapter invokes `prepare()` after serializing the exact wire messages, passes the request cancellation signal, rejects preparation or base-field collision before HTTP, merges the detached fields, and calls the captured `accept()` transaction after HTTP 2xx. The registry stops awaiting preparation after cancellation even if a contributor ignores the signal. Acceptance failures remain request failures under `REQUEST_EXTENSION`; transport and non-2xx failures never accept a contribution. A composition without the registry retains the reusable base adapter. Shipped compositions mount the registry and both contributors: package metadata is enabled by default, while Session-log upload is disabled by default and requires `session-log-deepseek.enabled: true`. Keyless `deepseek-official` replay invokes preparation with a synthetic empty base body and the same acceptance transaction before its first recorded chunk, preserving post-2xx extension side effects rather than field bytes.
`@deepseek-ai/dsh-deepseek-llm-api-extensions` registers `ctx.deepseekLlmApiExtensions`, an additive registry of top-level fields for `deepseek-official` request bodies. A contributor claims one declaration-merged field with `register()`. The adapter invokes `prepare()` after serializing the exact wire messages, passes the request cancellation signal, rejects preparation or base-field collision before HTTP, merges the detached fields, and calls the captured `accept()` transaction after HTTP 2xx. The registry stops awaiting preparation after cancellation even if a contributor ignores the signal. Acceptance failures remain request failures under `REQUEST_EXTENSION`; transport and non-2xx failures never accept a contribution. A composition without the registry retains the reusable base adapter. Shipped compositions mount the registry and both contributors: both package metadata and Session-log upload are enabled by default; `session-log-deepseek.enabled: false` disables log upload under the [default-upload decision](2026-09-14-session-log-upload-default.md). Keyless `deepseek-official` replay invokes preparation with a synthetic empty base body and the same acceptance transaction before its first recorded chunk, preserving post-2xx extension side effects rather than field bytes.
The provider-neutral `llm` package and `llm-pi-ai` contain no extension type, service lookup, field merge, or acceptance call.
## Incremental session-log field
`@deepseek-ai/dsh-session-log-deepseek` owns `dsh_session_log` as an explicit opt-in. When enabled, each request carrying a live Session id sends the contiguous canonical event suffix after the greatest durable `session-log-deepseek/delivery-accepted` watermark for that same Session identity. The field includes the immutable Session header and complete event envelopes. A 2xx appends a new watermark for the transmitted `throughSeq`; that event enters the following request's suffix. Forked logs retain parent watermark ids, so a child starts from sequence zero under its own identity. Concurrent acceptances may arrive out of order, and the maximum watermark remains authoritative. A process-local fold scans each Session event once and incrementally consumes later appends; a new Session object or HMR generation rebuilds the fold from durable history.
`@deepseek-ai/dsh-session-log-deepseek` owns the default-on `dsh_session_log` field. When enabled, each request carrying a live Session id sends the contiguous canonical event suffix after the greatest durable `session-log-deepseek/delivery-accepted` watermark for that same Session identity. The field includes the immutable Session header and complete event envelopes. A 2xx appends a new watermark for the transmitted `throughSeq`; that event enters the following request's suffix. Forked logs retain parent watermark ids, so a child starts from sequence zero under its own identity. Concurrent acceptances may arrive out of order, and the maximum watermark remains authoritative. A process-local fold scans each Session event once and incrementally consumes later appends; a new Session object or HMR generation rebuilds the fold from durable history.
The failure direction is at least once. A transport or provider rejection records no watermark. A crash after remote acceptance but before the watermark persists causes replay after resume, never a skipped sequence. Existing session checkpoints persist the event; the upload plugin owns no second store.
@@ -30,7 +30,7 @@ The `events` array contains complete canonical `SessionEvent` objects directly.
`@deepseek-ai/dsh-plugin-package-inventory-deepseek` owns the default-on `dsh_plugin_packages` field from the `llm` package family. It reads active non-group entries from the host Loader tree and, for a live requesting Agent, its standing preset tree. Node package resolution locates the owning manifest without requiring a `./package.json` export. Ordinary entries resolve from their owning tree, while a standing preset root mirrors its Loader's intentional harness-base override and nested includes retain their own bases. An anonymous nearest manifest marks a loose module; a named manifest must carry a version. Exact name/version pairs are deduplicated with deterministic ordering; simultaneously active versions remain separate.
Disabled, pending, failed, unloading, disposed, structural, loose non-package, ordinary dependency, programmatic child-fiber, and in-memory dynamic-plugin entries are outside this package inventory. This definition reports package-backed composition facts the runtime can prove instead of inventing provenance for arbitrary callbacks.
Disabled, pending, failed, unloading, disposed, structural, loose non-package, ordinary dependency, programmatic child-fiber, and in-memory dynamic-plugin entries are outside this package inventory. This definition reports package-backed composition facts the runtime can prove instead of inventing package identity for arbitrary callbacks.
## Deferred inventory caching
@@ -50,7 +50,7 @@ The process-lifetime manifest-identity cache remains separate because in-process
## Verification
Registry tests pin duplicate ownership, effect-scoped disposal, detached field values, concurrent and abortable preparation, receiver-preserving acceptance, one acceptance settlement, and failure aggregation. Session tests pin the default-off policy, explicit full-first/suffix-later delivery, direct complete event envelopes independent of base-body messages, incremental watermark folding, persisted restart recovery, fork identity fencing, out-of-order acceptance, and late invariant loading. Package-inventory tests pin default-on and explicit-off policies, host and standing-preset discovery, conflicting Loader resolution bases, manifest resolution, lifecycle filtering, and exact name/version ordering. The direct adapter mock proves pre-HTTP preparation failure, cancellation, non-2xx non-acceptance, 2xx acceptance before a later stream failure, and field collision. Keyless replay pins post-2xx extension acceptance, and the TypeScript JSON-RPC plus Python packaged-runtime snapshots project the acceptance event through both SDKs. Real Loader composition pins default package metadata plus opt-in Session upload, one real-API request mounts both shipped extensions and proves the official endpoint accepts them, and pi-ai tests retain their unchanged wire requests.
Registry tests pin duplicate ownership, effect-scoped disposal, detached field values, concurrent and abortable preparation, receiver-preserving acceptance, one acceptance settlement, and failure aggregation. Session tests pin default-on and explicit-off policies, full-first/suffix-later delivery, direct complete event envelopes independent of base-body messages, incremental watermark folding, persisted restart recovery, fork identity fencing, out-of-order acceptance, and late invariant loading. Package-inventory tests pin default-on and explicit-off policies, host and standing-preset discovery, conflicting Loader resolution bases, manifest resolution, lifecycle filtering, and exact name/version ordering. The direct adapter mock proves pre-HTTP preparation failure, cancellation, non-2xx non-acceptance, 2xx acceptance before a later stream failure, and field collision. Keyless replay pins post-2xx extension acceptance, and the TypeScript JSON-RPC plus Python packaged-runtime snapshots project the acceptance event through both SDKs. Real Loader composition pins default package metadata and Session upload plus explicit upload disablement, one real-API request mounts both shipped extensions and proves the official endpoint accepts them, and pi-ai tests retain their unchanged wire requests.
## Alternatives considered
@@ -73,11 +73,11 @@ The receiver would also need to traverse the tagged tree, resolve paths into the
### Why not omit assistant chunks or overlapping event data?
About 98% of the measured v1 real-session events were `assistant/chunk`. Omitting them after reference encoding reduced the complete identity JSON by another 84.79% for late enable and 6.49% for steady state, but prevented lossless reconstruction and left message provenance dangling. V2 embeds compact streams in attempt settlements; `dsh_session_log` still sends every current canonical event whole and does not omit those embedded records. Fuzzy or normalized substitutions have the same reconstruction defect.
About 98% of the measured v1 real-session events were `assistant/chunk`. Omitting them after reference encoding reduced the complete identity JSON by another 84.79% for late enable and 6.49% for steady state, but prevented lossless reconstruction and left message source-event references dangling. V2 embeds compact streams in attempt settlements; `dsh_session_log` still sends every current canonical event whole and does not omit those embedded records. Fuzzy or normalized substitutions have the same reconstruction defect.
**Keep the upload cursor only in memory.** Rejected because a normal process restart would resend the entire Session. A canonical acceptance event makes restart recovery best-effort durable without another storage backend; the remaining crash window produces allowed duplicates.
**Inventory every live Cordis fiber.** Rejected because programmatic and in-memory fibers have no authoritative npm package provenance. Loader-backed host and preset entries provide exact resolvable package identity.
**Inventory every live Cordis fiber.** Rejected because programmatic and in-memory fibers have no authoritative npm package identity. Loader-backed host and preset entries provide exact resolvable package identity.
**Cache one process-global list or expire it on a TTL.** Rejected because one immutable list is incorrect for Loader lifecycle and per-Session presets, while a TTL permits stale metadata between expiry boundaries. The deferred epoch design invalidates on the authoritative active-state transition instead.
@@ -85,7 +85,7 @@ About 98% of the measured v1 real-session events were `assistant/chunk`. Omittin
## Consequences
Official DeepSeek requests carry active package versions to their resolved `baseURL`, including configured gateways. An explicit Session-log opt-in also carries the complete newly unaccepted Session suffix. The fields are model-hidden and add no prompt tokens or KV-cache changes, but can substantially increase HTTP body size. Manifest resolution, field collision, acceptance logging, or provider schema rejection fails the model request rather than silently dropping metadata.
Official DeepSeek requests carry active package versions to their resolved `baseURL`, including configured gateways. Unless Session-log upload is disabled, each eligible request also carries the complete newly unaccepted Session suffix. The fields are model-hidden and add no prompt tokens or KV-cache changes, but can substantially increase HTTP body size. Manifest resolution, field collision, acceptance logging, or provider schema rejection fails the model request rather than silently dropping metadata.
The `delivery-accepted` event becomes part of the canonical log and is itself delivered on a later request. Crash recovery can duplicate a suffix but does not infer acceptance from assistant output or create a second local cursor store. Direct calls without a live Session omit the session field; host package inventory remains available.
@@ -14,13 +14,13 @@ Status: implemented
## 决策
`@deepseek-ai/dsh-deepseek-llm-api-extensions` 注册 `ctx.deepseekLlmApiExtensions`,即 `deepseek-official` 请求正文顶层字段的增量注册表。贡献方通过 `register()` 认领一个经声明合并的字段。适配器在序列化确切协议消息后调用 `prepare()`、传入请求取消信号,在 HTTP 前拒绝准备失败或基础字段冲突,合并分离字段,并在 HTTP 2xx 后调用捕获的 `accept()` 事务。即使贡献方忽略信号,注册表也会在取消后停止等待准备。接受失败仍以 `REQUEST_EXTENSION` 使请求失败;传输失败与非 2xx 失败绝不会接受贡献。未挂载注册表的组合会保留可复用基础适配器。随附组合会挂载注册表与两个贡献方:插件包元数据默认开启,会话日志上传默认关闭,需要设置 `session-log-deepseek.enabled: true`。无密钥 `deepseek-official` 回放会使用合成的空基础正文执行准备,并在第一个已记录分片前调用同一接受事务;它保持的是 2xx 后扩展副作用,而非字段字节。
`@deepseek-ai/dsh-deepseek-llm-api-extensions` 注册 `ctx.deepseekLlmApiExtensions`,即 `deepseek-official` 请求正文顶层字段的增量注册表。贡献方通过 `register()` 认领一个经声明合并的字段。适配器在序列化确切协议消息后调用 `prepare()`、传入请求取消信号,在 HTTP 前拒绝准备失败或基础字段冲突,合并分离字段,并在 HTTP 2xx 后调用捕获的 `accept()` 事务。即使贡献方忽略信号,注册表也会在取消后停止等待准备。接受失败仍以 `REQUEST_EXTENSION` 使请求失败;传输失败与非 2xx 失败绝不会接受贡献。未挂载注册表的组合会保留可复用基础适配器。随附组合会挂载注册表与两个贡献方:插件包元数据和会话日志上传均默认开启;按[默认上传决策](2026-09-14-session-log-upload-default.zh.md),设置 `session-log-deepseek.enabled: false` 可关闭日志上传。无密钥 `deepseek-official` 回放会使用合成的空基础正文执行准备,并在第一个已记录分片前调用同一接受事务;它保持的是 2xx 后扩展副作用,而非字段字节。
提供方无关的 `llm` 包与 `llm-pi-ai` 不包含任何扩展类型、服务查找、字段合并或接受调用。
## 增量会话日志字段
`@deepseek-ai/dsh-session-log-deepseek` 以显式选择启用的方式拥有 `dsh_session_log`。启用后,每个携带存活会话 id 的请求都会发送该确切会话身份最大持久 `session-log-deepseek/delivery-accepted` 水位之后的连续权威事件后缀。该字段包含不可变会话 header 与完整事件信封。2xx 会为已发送的 `throughSeq` 追加新水位;该事件会进入下一次请求的后缀。Fork 日志会保留父级水位 id,因此子会话会在自己的身份下从序列零开始。并发接受可能乱序到达,最大水位仍保持权威。进程内 fold 会让每条会话事件只被扫描一次,并增量消费后续追加;新的会话对象或 HMR generation 会从持久历史重建该 fold。
`@deepseek-ai/dsh-session-log-deepseek` 拥有默认开启的 `dsh_session_log` 字段。启用后,每个携带存活会话 id 的请求都会发送该确切会话身份最大持久 `session-log-deepseek/delivery-accepted` 水位之后的连续权威事件后缀。该字段包含不可变会话 header 与完整事件信封。2xx 会为已发送的 `throughSeq` 追加新水位;该事件会进入下一次请求的后缀。Fork 日志会保留父级水位 id,因此子会话会在自己的身份下从序列零开始。并发接受可能乱序到达,最大水位仍保持权威。进程内 fold 会让每条会话事件只被扫描一次,并增量消费后续追加;新的会话对象或 HMR generation 会从持久历史重建该 fold。
失败方向为至少一次。传输失败或提供方拒绝不会记录水位。远端接受后、水位持久化前发生崩溃,会在恢复后触发重放,绝不会跳过序列。现有会话检查点会持久化该事件;上传插件不拥有第二份存储。
@@ -50,7 +50,7 @@ Status: implemented
## 验证
注册表测试固定重复所有权、effect 作用域 dispose(资源释放)、分离字段值、并发且可取消的准备、保留接收者的接受操作、单次接受结算与失败聚合。会话测试固定默认关闭策略、显式启用后的首次完整/后续后缀交付、与基础正文消息无关的直接完整事件信封、增量水位 fold、持久化重启恢复、fork 身份围栏、乱序接受与 invariant 延迟加载。插件包清单测试固定默认开启与显式关闭策略、宿主与 standing preset 发现、冲突的 Loader 解析基址、manifest 解析、生命周期过滤及确切名称/版本排序。直接适配器 mock 测试证明 HTTP 前准备失败、取消、非 2xx 不接受、2xx 在后续流失败前接受,以及字段冲突。无密钥回放会固定 2xx 后扩展接受,TypeScript JSON-RPC 与 Python 打包运行时快照则通过两套 SDK 投影接受事件。真实 Loader 组合会固定默认包元数据与显式启用的会话上传,一个真实 API 请求会挂载两个随附扩展并证明官方端点接受它们;pi-ai 测试保持其协议请求不变。
注册表测试固定重复所有权、effect 作用域 dispose(资源释放)、分离字段值、并发且可取消的准备、保留接收者的接受操作、单次接受结算与失败聚合。会话测试固定默认开启与显式关闭策略、首次完整/后续后缀交付、与基础正文消息无关的直接完整事件信封、增量水位 fold、持久化重启恢复、fork 身份围栏、乱序接受与 invariant 延迟加载。插件包清单测试固定默认开启与显式关闭策略、宿主与 standing preset 发现、冲突的 Loader 解析基址、manifest 解析、生命周期过滤及确切名称/版本排序。直接适配器 mock 测试证明 HTTP 前准备失败、取消、非 2xx 不接受、2xx 在后续流失败前接受,以及字段冲突。无密钥回放会固定 2xx 后扩展接受,TypeScript JSON-RPC 与 Python 打包运行时快照则通过两套 SDK 投影接受事件。真实 Loader 组合会固定默认包元数据、默认会话上传与显式关闭上传,一个真实 API 请求会挂载两个随附扩展并证明官方端点接受它们;pi-ai 测试保持其协议请求不变。
## 考虑过的替代方案
@@ -73,7 +73,7 @@ Status: implemented
### 为什么不省略 assistant 分片或重叠事件数据?
实测 v1 真实 Session event 中约 98% 为 `assistant/chunk`。在引用编码后省略它们,会让完整 identity JSON 在延迟启用场景进一步减少 84.79%,在稳态场景进一步减少 6.49%,但会阻止无损重建并让 message provenance 悬空。V2 把紧凑 stream 嵌入 attempt settlement;`dsh_session_log` 仍会完整发送每个当前规范 event,且不会省略这些嵌入式 record。模糊或规范化替换也有相同重建缺陷。
实测 v1 真实 Session event 中约 98% 为 `assistant/chunk`。在引用编码后省略它们,会让完整 identity JSON 在延迟启用场景进一步减少 84.79%,在稳态场景进一步减少 6.49%,但会阻止无损重建并让 message source-event reference 悬空。V2 把紧凑 stream 嵌入 attempt settlement;`dsh_session_log` 仍会完整发送每个当前规范 event,且不会省略这些嵌入式 record。模糊或规范化替换也有相同重建缺陷。
**只在内存中保留上传游标。** 已否决,因为普通进程重启会重发完整会话。权威接受事件让重启恢复获得尽力而为的持久性,无需另一存储后端;剩余崩溃窗口只会产生允许的重复。
@@ -85,7 +85,7 @@ Status: implemented
## 后果
DeepSeek 官方请求会把存活包版本发送到解析后的 `baseURL`,包括已配置 gateway。显式选择启用会话日志后,请求还会携带完整的未接受会话新后缀。这些字段对模型不可见,不增加提示词 token,也不改变 KV Cache,但可能显著增大 HTTP 正文。Manifest 解析、字段冲突、接受记录或提供方 schema 拒绝会使模型请求失败,而不会静默丢弃元数据。
DeepSeek 官方请求会把存活包版本发送到解析后的 `baseURL`,包括已配置 gateway。除非关闭会话日志上传,否则符合条件的请求还会携带完整的未接受会话新后缀。这些字段对模型不可见,不增加提示词 token,也不改变 KV Cache,但可能显著增大 HTTP 正文。Manifest 解析、字段冲突、接受记录或提供方 schema 拒绝会使模型请求失败,而不会静默丢弃元数据。
`delivery-accepted` 事件会成为权威日志的一部分,并在后续请求中自行交付。崩溃恢复可能重复后缀,但不会根据 assistant 输出推断接受,也不会创建第二份本地游标存储。缺少存活会话的直接调用会省略会话字段;宿主包清单仍然可用。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-23-client-derived-tool-presentation.md
2026-08-23-client-derived-tool-presentation.md: 6b19dc881d572bfece345cbbd5eca688b2e05aab
2026-08-23-client-derived-tool-presentation.zh.md: 98ada31627398f317e4c41336af56e8c6cf0837e
2026-08-23-client-derived-tool-presentation.md: 703960cb24f90eb678fda381f929efb3f94bbd28
2026-08-23-client-derived-tool-presentation.zh.md: 7f161379553f3aca33c72c2edf93e114d27eb7f9
@@ -12,13 +12,13 @@ A `tool/result` does not repeat the tool name or arguments. Host-side result pre
Host projection would also duplicate structured data. Read, diff, search, and web results already persist bounded facts in `tool/result.data.meta`; another view object increases Remote payload size and Client decoding without adding durable meaning.
The Client already owns a complete tool-presentation entry point. `ui-chat` assembles `tool/call`, `tool/result`, and Code Dispatch events into stable `ToolCallBlock` values. `ui-tool` owns the recursive call tree, the `tool.call.toolview` keyed slot dispatched by tool name, the Generic fallback, card models, and details output. A business Client plugin can register a renderer for its own tool names.
The Client already owns a complete tool-presentation entry point. `ui-chat` assembles `tool/call`, `tool/result`, and PTC dispatch events into stable `ToolCallBlock` values. `ui-tool` owns the recursive call tree, the `tool.call.toolview` keyed slot dispatched by tool name, the Generic fallback, card models, and details output. A business Client plugin can register a renderer for its own tool names.
Splitting presentation between Host presenters and Client keyed renderers creates two interpretations of the same event. The keyed renderer is the Web extension point, so an intermediate Host view provides no independent Web capability.
`ToolDefinition.presentCall` and `presentResult` remain useful Host APIs even though ACP is automation-only and the repository has no production TUI consumer. Removing their definitions is a separate decision from keeping Session reads independent of presentation.
The required result is one raw Session journal and one Client presentation owner without visual degradation or incidental enhancement. Specialized cards, interactions, and Code Dispatch topology remain stable while the transport stops carrying transient views.
The required result is one raw Session journal and one Client presentation owner without visual degradation or incidental enhancement. Specialized cards, interactions, and PTC dispatch topology remain stable while the transport stops carrying transient views.
## Decision
@@ -26,7 +26,7 @@ The visual-equivalence requirements below exclude the separately approved [neste
The Session Remote journal sends only raw, validated, persistable Session events. `session.page` and `session.follow` do not parse tool arguments, query the Tools registry, restore a presenter scope, execute `presentCall` or `presentResult`, or construct or clone any tool view.
The Client Conversation layer continues to own tool call/result identity, pairing, lifecycle, Code Dispatch topology, and stable Chat Nodes. It does not interpret individual tool names or produce terminal, diff, read, search, or web component props.
The Client Conversation layer continues to own tool call/result identity, pairing, lifecycle, PTC dispatch topology, and stable Chat Nodes. It does not interpret individual tool names or produce terminal, diff, read, search, or web component props.
Client `ui-tool` continues to own card models and concrete renderers. Each card model directly reads the tool name, raw arguments, result content, error, durable metadata, Session cwd, and Host home from `ToolCallBlock`, and produces the same component props as the current page.
@@ -51,7 +51,7 @@ The Host `ToolDefinition.presentCall`, `ToolDefinition.presentResult`, `ToolCall
| Retained | the Session log format, Remote journal lifecycle, and Conversation identity/topology |
| Retained | the existing keyed slot, Generic fallback, and Chat, Details, and Trajectory structure |
| Forbidden | a new Client presenter service, parallel registry, or wire renderer id |
| Forbidden | new cards, visual redesign, interaction redesign, or Code Dispatch rich-card enhancements except the [nested terminal-card exception](../bug-fix/2026-09-05-nested-terminal-cards.md) |
| Forbidden | new cards, visual redesign, interaction redesign, or PTC dispatch rich-card enhancements except the [nested terminal-card exception](../bug-fix/2026-09-05-nested-terminal-cards.md) |
| Forbidden | compatibility dual-writing, version negotiation, or retention of the old `view` field |
## Terminology
@@ -96,7 +96,7 @@ The Host `ToolDefinition.presentCall`, `ToolDefinition.presentResult`, `ToolCall
1. The Client Session stores one contiguous raw event window.
2. `SessionEventSource` publishes `SessionEventEntry` values containing only events.
3. `ui-conversation` folds each event without a presentation companion.
4. The Chat and Trajectory Tool Definitions pair top-level calls and results by callId and assemble Code Dispatch subtrees.
4. The Chat and Trajectory Tool Definitions pair top-level calls and results by callId and assemble PTC dispatch subtrees.
5. `RunningToolCall` and `ToolResultNode` retain raw facts, metadata, and existing parent identity.
6. `ToolCallTree` dispatches `tool.call.toolview` by wire tool name.
7. `ui-tool` derives card component props from the block at the render site.
@@ -132,7 +132,7 @@ Session page/follow
Client SessionEventSource
-> Conversation Tool Definition
-> root call/result pairing + Code Dispatch topology
-> root call/result pairing + PTC dispatch topology
-> ToolCallBlock(name, argsRaw, content, error, meta)
-> tool.call.toolview keyed dispatch
-> Client card model
@@ -242,11 +242,11 @@ The Chat and Trajectory Tool Definitions read no views. They derive the followin
`ToolCallBlock` does not gain a generic `view`, `card`, `kind`, or `locations` field to replace the deleted fields. Concrete presentation remains the responsibility of `ui-tool` and keyed renderers.
### Root and Code Dispatch subcalls
### Root and PTC dispatch subcalls
Host presenter APIs describe top-level calls and results. Code Dispatch subcalls retain Generic, flattened presentation for the diff, read, search, and web models covered here; supported terminal calls use the same eligibility rules as roots.
Host presenter APIs describe top-level calls and results. PTC dispatch subcalls retain Generic, flattened presentation for the diff, read, search, and web models covered here; supported terminal calls use the same eligibility rules as roots.
Code Dispatch start and result events already carry `parentCallId`. Conversation preserves that existing fact on each child `ToolCallBlock`; root Session calls omit it. The diff, read, search, and web models accept only blocks without `parentCallId`; the terminal model and existing renderers that intentionally support nested calls accept child blocks.
PTC dispatch start and result events already carry `parentCallId`. Conversation preserves that existing fact on each child `ToolCallBlock`; root Session calls omit it. The diff, read, search, and web models accept only blocks without `parentCallId`; the terminal model and existing renderers that intentionally support nested calls accept child blocks.
Shared card models apply the same terminal eligibility and nonterminal child restrictions wherever a block renders, so no second presentation surface needs a placement field; the details panel that once delegated a selected block was removed with the right-hand details column ([decision](../feature/2026-09-04-right-sidebar-docking-infrastructure.md)).
@@ -308,7 +308,7 @@ The Client terminal model derives existing `TerminalBlock` props from the tool n
| settled persistent `bash`/`pwsh` | Generic flattened result, with no new exit card |
| foreground `terminal_send` | terminal prompt and output |
| background/error `terminal_send` | Generic result |
| Code Dispatch child | same terminal eligibility and fallback rules as a root call |
| PTC dispatch child | same terminal eligibility and fallback rules as a root call |
Standard shell results parse trailing `[exit code: N]` and `[killed by signal: X]` markers. A final recognized spill-policy notice selects Generic output instead: expandable in shell rows and raw in Details, because the exit marker may be displaced or omitted. A parsed marker is removed from the terminal body; timeout, sandbox denial, and markers without a pill remain in the body.
@@ -331,7 +331,7 @@ Standard and persistent providers sharing the same tool name are a special compa
| successful settled `write`/`edit` | applied contextual hunks from `meta.diffs` |
| settled `str_replace_editor` | Generic, because the tool defines no result presenter |
| write create or missing/malformed/empty applied metadata | current argument fallback |
| error, malformed arguments, edit with malformed metadata, or Code Dispatch child | Generic |
| error, malformed arguments, edit with malformed metadata, or PTC dispatch child | Generic |
Paths, `oldText:null`, `newText`, result-over-call diff precedence, the eight-line Chat limit, full-height Details presentation, and file-opening behavior remain unchanged.
@@ -339,7 +339,7 @@ Paths, `oldText:null`, `newText`, result-over-call diff precedence, the eight-li
A running `read` continues to show only the summary row. A successful settled `read` reads path, offset, lines, totalLines, and lang from result metadata and confirms that the result is one text block matching the read envelope.
Missing metadata, malformed fields, a mismatched result envelope, an error, a missing call head, or a Code Dispatch child all use Generic. Cwd-relative path labels, home abbreviation, syntax language, total line count, the eight-line Chat limit, and full-height Details presentation remain unchanged.
Missing metadata, malformed fields, a mismatched result envelope, an error, a missing call head, or a PTC dispatch child all use Generic. Cwd-relative path labels, home abbreviation, syntax language, total line count, the eight-line Chat limit, and full-height Details presentation remain unchanged.
The Client does not need to construct Host `ReadResultView.content`; Generic fallback can always read raw result content directly.
@@ -347,7 +347,7 @@ The Client does not need to construct Host `ReadResultView.content`; Generic fal
A running `grep` or `glob` continues to show only the argument summary. Successful results produce grouped matches or a path list from `meta.shape:'matches'` and `meta.shape:'paths'`, respectively.
The Client validates path, lineNumber, line, truncated, and total. Empty matches or paths form a valid card. Missing or malformed metadata, an unknown shape, an error, a missing call head, or a Code Dispatch child uses Generic.
The Client validates path, lineNumber, line, truncated, and total. Empty matches or paths form a valid card. Missing or malformed metadata, an unknown shape, an error, a missing call head, or a PTC dispatch child uses Generic.
When `truncated:true`, the card continues to show a recovery locator from raw result content. It does not show one when untruncated. The eight-line Chat limit, full-height Details presentation, and expansion behavior remain unchanged.
@@ -355,7 +355,7 @@ When `truncated:true`, the card continues to show a recovery locator from raw re
A running `web_search` or `web_fetch` continues to show only the summary row. A successful search builds the card from `meta.sources`, `meta.answer`, and `meta.truncated`; a successful fetch builds it from `meta.url`, `meta.statusCode`, and `meta.truncated`.
The Client validates every source's url, title, snippet, and publishedAt, and continues rendering only http/https URLs as links. Missing or malformed metadata, an error, a missing call head, or a Code Dispatch child uses Generic.
The Client validates every source's url, title, snippet, and publishedAt, and continues rendering only http/https URLs as links. Missing or malformed metadata, an error, a missing call head, or a PTC dispatch child uses Generic.
Search answer text, source ordering, label fallback, and truncation notice remain unchanged. The fetch final URL, status, truncation notice, and raw body below Details remain unchanged.
@@ -387,7 +387,7 @@ This change does not add a general tool-side-effect registry. The ability for a
## Fixtures and Test Data
The Client fixture deletes its handwritten `presentCall()`, `presentResult()`, `viewFor()`, and fixture tool-view types. It continues producing the same raw calls, result content, and result metadata as a real log.
The assembled RemoteMock scenario contains no handwritten `presentCall()`, `presentResult()`, `viewFor()`, or tool-view types. It supplies the same raw calls, result content, and result metadata as a real log.
| Fixture | Raw facts that must remain |
|---|---|
@@ -398,7 +398,7 @@ The Client fixture deletes its handwritten `presentCall()`, `presentResult()`, `
| web | result metadata sources/answer or url/statusCode/truncated |
| generic/custom | name, argsRaw, content, and error |
The fixture does not import Host tool packages to compute page presentation and retains no presenter mirror. The same raw fixture continues to drive jsdom, built Web snapshots, and the `?fixture` browser path.
The scenario does not import Host tool packages to compute page presentation and retains no presenter mirror. The same raw scenario drives built Web snapshots under jsdom; real-Host browser cases independently cover the network path.
## Presentation-Equivalence Matrix
@@ -420,7 +420,7 @@ The fixture does not import Host tool packages to compute page presentation and
| grep/glob | current grouped/path card, truncation, and recovery |
| web_search/web_fetch | current source/summary card and raw body |
| Todo/Question/Skill/Cordis | current specialized rows |
| Code Dispatch subcall | terminal cards when eligible; diff/read/search/web remain Generic/flattened |
| PTC dispatch subcall | terminal cards when eligible; diff/read/search/web remain Generic/flattened |
| Chat and Details | identical card fields for the same call |
| Trajectory | current identity, tree, selection, and details |
| Deliverables | current successful-mutation chips and links |
@@ -478,7 +478,7 @@ This change does not promise to preserve differences expressed only through a Ho
- Conversation input and Tool blocks contain no view fields.
- Chat and Trajectory Tool Definitions read raw events.
- Event pairing, Context replay, trees, and target snapshots remain unchanged.
- Child Tool blocks preserve the existing Code Dispatch `parentCallId`; row and Details slot owner props add no separate placement field.
- Child Tool blocks preserve the existing PTC dispatch `parentCallId`; row and Details slot owner props add no separate placement field.
### UI Tool and Deliverables
@@ -515,7 +515,7 @@ This change does not promise to preserve differences expressed only through a Ho
- replace, prepend, and append accept entries without views.
- Chat and Trajectory root call/result pairing remains unchanged.
- The Code Dispatch tree remains unchanged.
- The PTC dispatch tree remains unchanged.
- Result-only fallback remains unchanged.
- A synthetic interruption result copies no view.
- Node identity across registry rebuild, older prepend, and live append remains unchanged.
@@ -566,7 +566,7 @@ Changes to this decision use `dsh-pre-push-checks` to select commands for the fi
- ui-chat and ui-trajectory Tool Definition tests;
- ui-tool terminal, diff, read, search, web, row, tree, and details tests;
- ui-deliverables produced-file tests;
- connection fixture and Client runtime tests;
- assembled RemoteMock and Client runtime tests;
- affected Host and Client TypeScript faces;
- lint and duplication;
- per-file 100% coverage for affected source files;
@@ -589,7 +589,7 @@ Changes to this decision use `dsh-pre-push-checks` to select commands for the fi
- Deliverables does not depend on render intent and preserves current paths.
- Text, components, expanded content, states, links, and ordering for all first-party top-level tools remain unchanged.
- Malformed, missing-metadata, error, orphan, and unknown-tool cases continue to fall back safely.
- Code Dispatch diff, read, search, and web subcalls remain Generic and flattened; terminal subcalls follow root eligibility.
- PTC dispatch diff, read, search, and web subcalls remain Generic and flattened; terminal subcalls follow root eligibility.
- Chat, Details, and Trajectory behavior remains unchanged.
- Existing Web browser expected outputs pass without refresh.
- Host presenter APIs, implementations, and direct tests remain unchanged.
@@ -634,7 +634,7 @@ An on-demand RPC would turn one page read into N network calls and would still r
### Allow presentation enhancements
Bundling richer Code Dispatch cards, missing-call-head inference, or other historical presentation enhancements with the ownership change would prevent snapshots from proving equivalence. This decision rejects that coupling; the [nested terminal-card exception](../bug-fix/2026-09-05-nested-terminal-cards.md) does not relax nonterminal child restrictions.
Bundling richer PTC dispatch cards, missing-call-head inference, or other historical presentation enhancements with the ownership change would prevent snapshots from proving equivalence. This decision rejects that coupling; the [nested terminal-card exception](../bug-fix/2026-09-05-nested-terminal-cards.md) does not relax nonterminal child restrictions.
### Accept temporary Generic degradation
@@ -703,7 +703,7 @@ This note preserves result metadata from the [canonical tool output contract](20
## Deferred
- A separate explicit decision may evaluate deleting Host presenters if they remain without production consumers; this decision does not prejudge it.
- Specialized diff, read, search, and web cards for Code Dispatch subcalls require a separate design and visible-snapshot updates; terminal calls are covered by the linked partial supersession.
- Specialized diff, read, search, and web cards for PTC dispatch subcalls require a separate design and visible-snapshot updates; terminal calls are covered by the linked partial supersession.
- A third-party mutation tool that joins Deliverables requires a new Client-owned contribution; this decision does not create a registry for an absent consumer.
- Distinct Client presentation for same-named providers first requires a stable, non-presentational identity; it must not restore per-page Host views.
- If Client card-model performance needs measurement, an immutable-block microbenchmark can be added; the shipped architecture already prohibits scanning the Session window.
@@ -12,13 +12,13 @@ Session 历史是持久 journal 接口,工具卡片属于 Client 展示。在
Host 投影还会重复结构化数据。read、diff、search 与 web 结果已在 `tool/result.data.meta` 中持久化有界事实;另一份 view 只增加 Remote payload 与 Client 解码成本,不增加持久语义。
Client 已经拥有完整的工具展示入口。`ui-chat` 将 `tool/call`、`tool/result` 与 Code Dispatch 事件组装成稳定的 `ToolCallBlock`;`ui-tool` 拥有递归调用树、按工具名称分发的 `tool.call.toolview` keyed slot、Generic fallback、卡片模型和 details output;业务 Client 插件可以为自己的工具名称注册 renderer。
Client 已经拥有完整的工具展示入口。`ui-chat` 将 `tool/call`、`tool/result` 与 PTC dispatch 事件组装成稳定的 `ToolCallBlock`;`ui-tool` 拥有递归调用树、按工具名称分发的 `tool.call.toolview` keyed slot、Generic fallback、卡片模型和 details output;业务 Client 插件可以为自己的工具名称注册 renderer。
Host presenter 与 Client keyed renderer 分担展示会形成对同一事件的两套解释。keyed renderer 是 Web 扩展点,因此中间 Host view 不提供独立 Web 能力。
`ToolDefinition.presentCall`/`presentResult` 仍是保留的 Host API;ACP 采用 automation-only 协议,仓库也没有生产 TUI consumer。是否删除这些定义与 Session 读取是否独立于展示是两个决定。
所需结果是一条原始 Session journal 和一个 Client 展示 owner,且不发生可见退化或顺带增强。专用卡片、交互和 Code Dispatch 拓扑保持稳定,transport 不再携带临时 view。
所需结果是一条原始 Session journal 和一个 Client 展示 owner,且不发生可见退化或顺带增强。专用卡片、交互和 PTC dispatch 拓扑保持稳定,transport 不再携带临时 view。
## Decision
@@ -26,7 +26,7 @@ Host presenter 与 Client keyed renderer 分担展示会形成对同一事件的
Session Remote journal 只下发原始、已验证、可持久化的 Session event。`session.page` 和 `session.follow` 不解析工具参数,不查询 Tools registry,不恢复 presenter scope,不执行 `presentCall`/`presentResult`,也不构造或克隆任何 tool view。
Client Conversation 层继续负责工具调用与结果的 identity、配对、生命周期、Code Dispatch 拓扑和稳定 Chat Node。它不解释具体工具名称,也不生成 terminal、diff、read、search 或 web 组件 props。
Client Conversation 层继续负责工具调用与结果的 identity、配对、生命周期、PTC dispatch 拓扑和稳定 Chat Node。它不解释具体工具名称,也不生成 terminal、diff、read、search 或 web 组件 props。
Client `ui-tool` 继续负责 card model 和具体 renderer。每个 card model 改为直接读取 `ToolCallBlock` 中的工具名称、原始参数、结果内容、错误、持久 metadata、Session cwd 与 Host home,并生成与现有页面相同的组件 props。
@@ -51,7 +51,7 @@ Host 的 `ToolDefinition.presentCall`、`ToolDefinition.presentResult`、`ToolCa
| 保留 | Session 日志格式、Remote journal 生命周期与 Conversation identity/topology |
| 保留 | 现有 keyed slot、Generic fallback、Chat、Details 与 Trajectory 结构 |
| 禁止 | 新 Client presenter service、平行 registry 或 wire renderer id |
| 禁止 | 新卡片、视觉改版、交互改版或 Code Dispatch rich-card 增强,[嵌套 terminal 卡片例外](../bug-fix/2026-09-05-nested-terminal-cards.zh.md)除外 |
| 禁止 | 新卡片、视觉改版、交互改版或 PTC dispatch rich-card 增强,[嵌套 terminal 卡片例外](../bug-fix/2026-09-05-nested-terminal-cards.zh.md)除外 |
| 禁止 | 为兼容保留双写、版本协商或旧 `view` 字段 |
## 术语
@@ -96,7 +96,7 @@ Host 的 `ToolDefinition.presentCall`、`ToolDefinition.presentResult`、`ToolCa
1. Client Session 保存一个连续 raw event window。
2. `SessionEventSource` 发布只含 event 的 `SessionEventEntry`。
3. `ui-conversation` 在没有 presentation companion 的情况下 fold 每个事件。
4. Chat 与 Trajectory Tool Definition 按 callId 配对顶层 call/result,并组装 Code Dispatch 子树。
4. Chat 与 Trajectory Tool Definition 按 callId 配对顶层 call/result,并组装 PTC dispatch 子树。
5. `RunningToolCall` 与 `ToolResultNode` 保存 raw facts、metadata 与既有 parent identity。
6. `ToolCallTree` 按 wire tool name 分发 `tool.call.toolview`。
7. `ui-tool` 在 render site 从 block 派生 card component props。
@@ -132,7 +132,7 @@ Session page/follow
Client SessionEventSource
-> Conversation Tool Definition
-> root call/result pairing + Code Dispatch topology
-> root call/result pairing + PTC dispatch topology
-> ToolCallBlock(name, argsRaw, content, error, meta)
-> tool.call.toolview keyed dispatch
-> Client card model
@@ -242,11 +242,11 @@ Chat 和 Trajectory 的 Tool Definition 都不读取 view,而从事件生成
`ToolCallBlock` 不新增通用 `view`、`card`、`kind` 或 `locations` 字段替代被删除字段。具体展示仍只属于 `ui-tool` 与 keyed renderer。
### Root 与 Code Dispatch 子调用
### Root 与 PTC dispatch 子调用
Host presenter API 描述顶层 call/result。本决定覆盖的 diff、read、search 和 web model 对 Code Dispatch 子调用保留 Generic/flattened 展示;受支持的 terminal 调用使用与根调用相同的适用规则。
Host presenter API 描述顶层 call/result。本决定覆盖的 diff、read、search 和 web model 对 PTC dispatch 子调用保留 Generic/flattened 展示;受支持的 terminal 调用使用与根调用相同的适用规则。
Code Dispatch start 与 result event 已经携带 `parentCallId`。Conversation 在每个 child `ToolCallBlock` 上保留这项现有事实,root Session call 则不携带它。diff、read、search 和 web model 只接受没有 `parentCallId` 的 block;terminal model 与原本有意支持嵌套调用的 renderer 接受 child block。
PTC dispatch start 与 result event 已经携带 `parentCallId`。Conversation 在每个 child `ToolCallBlock` 上保留这项现有事实,root Session call 则不携带它。diff、read、search 和 web model 只接受没有 `parentCallId` 的 block;terminal model 与原本有意支持嵌套调用的 renderer 接受 child block。
共享的 card model 在 block 渲染到哪里都施加同样的终端资格与非终端子调用限制,因此不需要第二个展示面带 placement 字段;曾经原样委托选中 block 的详情面板已随右侧详情列一并删除([决策](../feature/2026-09-04-right-sidebar-docking-infrastructure.zh.md))。
@@ -308,7 +308,7 @@ Client terminal model 从工具名称、调用参数、结果 content、error
| persistent `bash`/`pwsh` settled | Generic flattened result,不新增 exit card |
| `terminal_send` 前台 | terminal prompt 与 output |
| `terminal_send` background/error | Generic 结果 |
| Code Dispatch child | 与根调用相同的 terminal 适用规则与 fallback 规则 |
| PTC dispatch child | 与根调用相同的 terminal 适用规则与 fallback 规则 |
标准 shell 结果解析末尾 `[exit code: N]` 与 `[killed by signal: X]`。末尾已识别的 spill 策略提示会改用 Generic 输出:在 shell 行中可展开,在 Details 中显示原文,因为退出标记可能被移位或省略。已解析的 marker 从 terminal 正文移除;timeout、sandbox denial 与没有 pill 的 marker 留在正文。
@@ -331,7 +331,7 @@ TerminalBlock 的 ANSI、光标重放、宽字符、行数上限、展开、复
| settled `write`/`edit` success | 从 `meta.diffs` 生成 applied contextual hunks |
| settled `str_replace_editor` | Generic,因为该工具没有 result presenter |
| write create 或 applied metadata 缺失、畸形、为空 | 当前 args fallback |
| error、畸形 args、edit 的 metadata 畸形、Code Dispatch child | Generic |
| error、畸形 args、edit 的 metadata 畸形、PTC dispatch child | Generic |
路径、`oldText:null`、`newText`、结果覆盖调用时 diff、Chat 8 行上限、Details 全高显示和文件打开行为不变。
@@ -339,7 +339,7 @@ TerminalBlock 的 ANSI、光标重放、宽字符、行数上限、展开、复
running `read` 继续只有摘要行。成功 settled `read` 从 result meta 读取 path、offset、lines、totalLines 与 lang,并确认结果是单个文本块且符合 read envelope。
meta 缺失、字段畸形、result envelope 不匹配、error、缺失 call head 或 Code Dispatch child 都走 Generic。路径 label 的 cwd 相对化、home 缩写、语法语言、总行数、Chat 8 行上限与 Details 全高显示不变。
meta 缺失、字段畸形、result envelope 不匹配、error、缺失 call head 或 PTC dispatch child 都走 Generic。路径 label 的 cwd 相对化、home 缩写、语法语言、总行数、Chat 8 行上限与 Details 全高显示不变。
Client 不需要构造 Host `ReadResultView.content`;Generic fallback 始终可直接读取原始 result content。
@@ -347,7 +347,7 @@ Client 不需要构造 Host `ReadResultView.content`;Generic fallback 始终
running `grep`/`glob` 继续只有参数摘要。成功结果分别从 `meta.shape:'matches'` 与 `meta.shape:'paths'` 生成 grouped matches 或 path list。
Client 校验 path、lineNumber、line、truncated 与 total。空 matches/paths 是有效卡片;缺失/畸形 meta、未知 shape、error、缺失 call head 与 Code Dispatch child 走 Generic。
Client 校验 path、lineNumber、line、truncated 与 total。空 matches/paths 是有效卡片;缺失/畸形 meta、未知 shape、error、缺失 call head 与 PTC dispatch child 走 Generic。
`truncated:true` 时继续从原始 result content 显示 recovery locator;未截断时不显示。Chat 8 行上限、Details 全高显示和展开行为不变。
@@ -355,7 +355,7 @@ Client 校验 path、lineNumber、line、truncated 与 total。空 matches/paths
running `web_search`/`web_fetch` 继续只有摘要行。成功 search 从 `meta.sources`、`meta.answer`、`meta.truncated` 生成卡片;成功 fetch 从 `meta.url`、`meta.statusCode`、`meta.truncated` 生成卡片。
Client 校验每个 source 的 url、title、snippet 与 publishedAt,并继续只把 http/https URL 渲染为链接。meta 缺失或畸形、error、缺失 call head 与 Code Dispatch child 走 Generic。
Client 校验每个 source 的 url、title、snippet 与 publishedAt,并继续只把 http/https URL 渲染为链接。meta 缺失或畸形、error、缺失 call head 与 PTC dispatch child 走 Generic。
search 的 answer、来源顺序、label fallback 与截断提示不变;fetch 的最终 URL、状态、截断提示与 Details 下方原始正文不变。
@@ -387,7 +387,7 @@ Deliverables Definition 按 callId 观察 root `tool/call` 与成功 `tool/resul
## Fixture 与测试数据
Client fixture 删除手写 `presentCall()`、`presentResult()`、`viewFor()` 与 fixture tool-view 类型。它继续产生与真实日志相同的 raw call、result content 和 result meta。
组装 RemoteMock 场景不包含手写 `presentCall()`、`presentResult()`、`viewFor()` 或 tool-view 类型。它提供与真实日志相同的 raw call、result content 和 result meta。
| Fixture | 必须保留的原始事实 |
|---|---|
@@ -398,7 +398,7 @@ Client fixture 删除手写 `presentCall()`、`presentResult()`、`viewFor()`
| web | result meta 的 sources/answer 或 url/statusCode/truncated |
| generic/custom | name、argsRaw、content、error |
fixture 不导入 Host 工具包来计算页面展示,也不保留 presenter 镜像。同一 raw fixture 继续驱动 jsdom、built Web snapshot 与 `?fixture` 浏览器路径。
该场景不导入 Host 工具包来计算页面展示,也不保留 presenter 镜像。同一 raw 场景在 jsdom 下驱动 built Web snapshot;真实 Host 浏览器用例独立覆盖网络路径。
## 展示等价矩阵
@@ -420,7 +420,7 @@ fixture 不导入 Host 工具包来计算页面展示,也不保留 presenter
| grep/glob | 当前 grouped/path card、截断与 recovery |
| web_search/web_fetch | 当前来源/摘要 card 与原始正文 |
| Todo/Question/Skill/Cordis | 当前专用行 |
| Code Dispatch subcall | 满足条件时显示 terminal 卡片;diff/read/search/web 保持 Generic/flattened |
| PTC dispatch subcall | 满足条件时显示 terminal 卡片;diff/read/search/web 保持 Generic/flattened |
| Chat 与 Details | 同一调用使用相同 card fields |
| Trajectory | 当前 identity、树、选择和 details |
| Deliverables | 当前成功 mutation chips 与链接 |
@@ -478,7 +478,7 @@ Host registry 允许不同 scope 为同一 tool name 提供不同定义;Sessio
- Conversation input 与 Tool block 不含 view 字段。
- Chat/Trajectory Tool Definition 读取 raw event。
- event pairing、Context replay、树与 target snapshot 保持不变。
- child Tool block 保留现有 Code Dispatch `parentCallId`;row 与 Details slot owner props 都不增加独立 placement 字段。
- child Tool block 保留现有 PTC dispatch `parentCallId`;row 与 Details slot owner props 都不增加独立 placement 字段。
### UI Tool 与 Deliverables
@@ -515,7 +515,7 @@ Host registry 允许不同 scope 为同一 tool name 提供不同定义;Sessio
- replace、prepend 与 append 接受无 view entry。
- Chat 与 Trajectory root call/result 配对不变。
- Code Dispatch 树不变。
- PTC dispatch 树不变。
- result-only fallback 不变。
- interruption synthetic result 不复制 view。
- registry rebuild、older prepend 与 live append 的 Node identity 不变。
@@ -566,7 +566,7 @@ Host registry 允许不同 scope 为同一 tool name 提供不同定义;Sessio
- ui-chat 与 ui-trajectory Tool Definition 测试;
- ui-tool terminal、diff、read、search、web、row、tree 与 details 测试;
- ui-deliverables produced-files 测试;
- connection fixture 与 Client runtime 测试;
- 组装 RemoteMock 与 Client runtime 测试;
- 受影响 Host/Client TypeScript face;
- lint 与 duplication;
- 受影响源文件 per-file 100% coverage;
@@ -589,7 +589,7 @@ Host registry 允许不同 scope 为同一 tool name 提供不同定义;Sessio
- Deliverables 不依赖 render intent 且保持当前 paths。
- 所有第一方顶层工具的文本、组件、展开内容、状态、链接与排序不变。
- malformed、missing-meta、error、orphan 与 unknown-tool 继续安全 fallback。
- Code Dispatch 的 diff、read、search 和 web 子调用保持 Generic/flattened;terminal 子调用遵循根调用适用规则。
- PTC dispatch 的 diff、read、search 和 web 子调用保持 Generic/flattened;terminal 子调用遵循根调用适用规则。
- Chat、Details 与 Trajectory 行为不变。
- 现有 Web browser expected 无需刷新即可通过。
- Host presenter API、实现与直接测试不变。
@@ -634,7 +634,7 @@ read 行结构、applied diff、search 分组、web sources 和有效 truncation
### 允许展示增强
将更丰富的 Code Dispatch 卡片、缺失 call head 的推断或其他历史展示增强与所有权变更捆绑,会使快照无法证明对等。本决定拒绝这种捆绑;[嵌套 terminal 卡片例外](../bug-fix/2026-09-05-nested-terminal-cards.zh.md)不放宽非 terminal 子调用限制。
将更丰富的 PTC dispatch 卡片、缺失 call head 的推断或其他历史展示增强与所有权变更捆绑,会使快照无法证明对等。本决定拒绝这种捆绑;[嵌套 terminal 卡片例外](../bug-fix/2026-09-05-nested-terminal-cards.zh.md)不放宽非 terminal 子调用限制。
### 接受临时 Generic 退化
@@ -703,7 +703,7 @@ optional `view` 的缺失是所有 consumer 共同遵守的预发布 wire 类型
## Deferred
- Host presenter 若长期没有生产消费者,可由另一项明确决策评估删除;本决定不预判。
- Code Dispatch 子调用的 diff、read、search 和 web 专用卡片仍需独立设计并更新可见快照;terminal 调用由链接的部分取代决策负责。
- PTC dispatch 子调用的 diff、read、search 和 web 专用卡片仍需独立设计并更新可见快照;terminal 调用由链接的部分取代决策负责。
- 第三方 mutation tool 若要加入 Deliverables,需新增 Client-owned 贡献;本决定不为尚无消费者的扩展性建 registry。
- 同名 provider 若要不同 Client 展示,需先定义稳定、非展示性的 identity;不得恢复按页 Host view。
- Client card model 若需量化性能,可以增加 immutable-block 微基准;已交付架构禁止扫描 Session window。
@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write .agents/notes/implemented/architecture/2026-08-23-locale-owned-client-ui-copy.md
2026-08-23-locale-owned-client-ui-copy.md: 623189b8faa45f0ef2a1b724d3e050d33f7d59bd
2026-08-23-locale-owned-client-ui-copy.zh.md: e2a6f49de89e6e6ba2734dd69cf536a3e13d2aea
2026-08-23-locale-owned-client-ui-copy.md: bc4d8efbce1f246f712e3afe6af40d1518e8cb17
2026-08-23-locale-owned-client-ui-copy.zh.md: ce996a4aafc392e739bec2eccd8714cfeff7d8e9
@@ -12,6 +12,8 @@ Typed locale namespaces and bilingual dictionary parity proved that registered d
**Locale dictionaries own all product-authored client UI wording.** Visible text, accessibility names, tooltips, placeholders, empty states, status labels, units, and formatting templates reach presentation through a typed `t` seat or an already-localized prop. A value authored by a user, model, provider, plugin, wire peer, or operating system remains data and renders verbatim; protocol tags, tool names, paths, URLs, JSON/JavaScript literals, and stable internal ids are not translated.
Product-owned catalog descriptions follow the same rule. The client maps an exact built-in provider, model, and description to a locale key; a changed description or an external provider description remains provider data and renders verbatim.
**Cordis-free primitives require complete localized copy props and own no language fallback.** `MarkdownText`, `JsonTree`, `TerminalBlock`, `DiffBlock`, `ReadBlock`, `SearchBlock`, `WebBlock`, `CodeBlock`, `JsonBlock`, `HoverCard`, and `ConnectionIndicator` receive their chrome from the feature render site. This preserves the primitive package's runtime independence while making omission a type error instead of silently selecting Chinese or English. Shared words live in the `common` namespace; feature-specific phrases stay with the feature that decides their meaning.
**Localized display text is never an identity.** Models and stores retain discriminants, stable ids, and non-display markers. Renderers translate after matching, and request maps carry stable group membership into the trajectory ledger. A client-synthesized error that must survive in a view model uses a stable marker and is translated only when displayed. Language switching therefore changes wording without changing selection, grouping, search identity, or lifecycle state.
@@ -22,7 +24,7 @@ The product-authored error and design-literal exclusions, primitive defaults, an
## Verification
The AST check's own Vitest spec pins direct JSX, template branches, semantic copy props, label data, formatter returns, locale-key calls, structural attributes, and dictionary owners. Locale dictionary parity pins identical `zh`/`en` keys. Client component suites exercise both direct translated seats and locale-prop adapters, and the assembled web replay plus the required real-server GIF demonstrate the shipped locale switch on the actual trajectory surface.
The AST check's own Vitest spec pins direct JSX, template branches, semantic copy props, label data, formatter returns, locale-key calls, structural attributes, and dictionary owners. Locale dictionary parity pins identical `zh`/`en` keys. Client suites exercise direct translated seats, locale-prop adapters, and built-in catalog-description localization without altering external descriptions. The assembled web replay plus the required real-server GIF demonstrate the shipped locale switch on the actual trajectory surface.
## Alternatives considered
Loaded 100 of 4570 files, more files were not shown because too many files have changed in this diff. Show more