Keeps the apply_patch toolview registration with master's file-menu inject,
takes master's shortened slot JSDoc, re-adds the seven filesystem type-equiv
manifest entries, regenerates the event matrix and client catalogs, moves the
fs-observation-policy upgrade guide under v0.2.2-alpha.2 with the matching
fs-patch version, and refreshes the two SDK scenarios master added for the
write and edit guidance change.
Evaluation on the shipped Web composition decided the defaults: the
`fs-observation-policy` row in dsh-base ships `disabled: true`, so write,
edit, and apply_patch mutate unconditionally in every profile; the Web
`standard` preset lists apply_patch; and the tool-fs `applyPatchGuard` switch
is gone, apply_patch always consulting the fs/* intent waterfalls like the
other mutating tools.
The write and edit guidance stops citing the policy as the default, and the
apply_patch guidance asks for enough hunk context instead of a prior read.
The two recordings that capture a policy refusal (fs-policy-reject and
apply-patch-rejected) re-enable the row through a per-scenario
runtime.cordis.yml; every other sidecar follows the prompt change, and the
Web standard-preset recordings gain the apply_patch schema. An upgrade guide
records the default change.
`tool-fs.applyPatchGuard` selects the mutation guards of apply_patch:
`observation` consults the mounted observation policy per operation like
write and edit; `none`, the default, applies hunks unconditionally on
context match alone as the Codex tool does, keeps publishing fs/observed,
and replaces the read-first guidance with a context-matching reminder.
In 2,226 recorded DS-Bench and DeepSWE rollouts with the policy on, read-first
and stale-version refusals were 29–41% of apply_patch failures although hunk
context matching already rejects a divergent file. The recorded apply-patch
scenarios pin `observation`, under which they were captured.
Across 2,226 recorded DS-Bench and DeepSWE rollouts, 14% of apply_patch
calls arrived as {"input": ...} and were rejected for the missing `patch`
property: models trained on the Codex tool send its parameter name
unprompted. The schema now uses `input`; the executor request keeps its
internal `patch` field. Recorded apply-patch sessions, the schema sidecar,
and the catalogs follow.
Takes master's re-measured npm install-layout budget, which covers this
branch's graph, and regenerates the Cordis API, event matrix, client slot,
and tool catalogs on the merged tree.
Resolves the headless snapshot runner (keeps both the readonly-patch-directory
and cua-script-guidance workspace setups), regenerates the client slot and tool
catalogs with their pairing record, adds the apply_patch tool title to the five
locales master introduced, and refreshes the apply-patch system-prompt pin for
master's subagent guidance change.
Restore files outside the patch feature to their merge-base state: the
invariant-check refactors across acp, session-controller, client,
compaction, agent-loop, agent, session-title, and subprocess, the cast
baseline edit, and the rescope-vendor skip list added during the merge.
Rename tool-fs.enableApplyPatch to applyPatch, matching the read, write,
and edit switches. Remove the unused collectMutationPaths export and the
unconsumed PatchHunk.contextLineIndices field, including its SSH schema
and fixtures. Report preflight refusals and syntax failures directly
instead of padding every phase with outcome, affectedPaths, and
cancelled fields. Add apply_patch to the shipped tool catalog test.
Fold the default-activation Agent Note into the apply_patch file-editing
note and delete it. Delete the guarded-filesystem-mutations upgrade
guide: every change is additive and all FileSystem subclasses live in
this repository. Trim defensive and reviewer-addressed sentences from
the package READMEs, subsystem docs, FileSystem JSDoc, and the Agent
Note, keeping the English and Chinese pairs aligned.
Regenerate config/tool/module/event catalogs, Cordis API and client slot catalogs; re-record reviewed bilingual hashes instead of merging generated bytes.
Accept deleted filesystem invariant companions. Master has no replacement event-invariant registration; keep delete-intent in typed events, policy guards, and behavior tests.
Adopt master workspace-changes intent waterfalls and tool-execution recorder association, add a prepended delete-intent listener that delegates before capture, remove argument parsing and the fs-patch dependency, and retain the real-tool Review matrix.
Keep master filesystem mutation-result metadata and tool switches alongside opt-in apply_patch. Await current-directory resolution in plan and no-follow preflight; store patch diffs as processPath. Adapt native/PTC fixtures to the supported modes.
Preserve client patch diffs and stopped states with master partial arguments and absolute navigation. Use master's rewritten ProviderEditor, including its remaining repeated model validation.
Regenerate the lockfile with pnpm install and align the new fs-patch package version. Clean retired package build outputs. Update exact rescope exceptions for verified preset and event identifiers from master; old package names remain rejected.
Delete @deepseek-ai/dsh-invariants, every package's ./invariant companion
and its tests, the invariant gates and Vitest setup, the scoped-event
resolver generator, and the INVARIANT-coded listener rethrow paths.
Update docs, READMEs, Agent Notes, and add an upgrade guide.
* feat(i18n): key translation pairing records by heading section
Replace the whole-file blob hashes in *.i18n.yaml with one entry per
heading section that still holds language-specific content, keyed by the
English heading-slug path. Blocks identical across the pair (generated
table rows, code blocks, shared regions) stay out of the hash, so edits to
different sections and identical regenerations merge with Git's default
text merge, including on GitHub.
Remove the dsh-translation-pairing merge driver, its resolver command and
installer wiring; gen-translation-brief recovers last-confirmed text from
commit history. Migrate all active records, archive the superseded Agent
Note, and record the replayed conflict reduction in the new one.
* test(i18n): refresh translation prompt snapshot for the terminology note
* feat(i18n): exclude only code blocks and generated regions from pairing records
Replace byte-equality masking with an explicit rule: a record hash skips
fenced code blocks and BEGIN/END GENERATED regions, the two kinds of
content the pairing gate already requires to be identical on both sides.
Move generated data into regions written into both pages:
gen-config-catalog renders both languages and the record with per-package
regions labelled by code identifiers; gen-module-graph keeps its table in a
region; gen-doc-graphs splices the event matrix regions into the Chinese
page. Add a generic region splice beside the region grammar.
Regenerate the client and persistence catalogs, document the review address builder with a plain parameter, share the changed-file coordinate parsing between the comparison and native-open routes, and fold the summary and comparison stores onto one Host read store. A path with no existing ancestor but the filesystem root keeps its spelling on every platform, and the temp-root test redirects TMP and TEMP as well as TMPDIR while counting only the recorder's own directories.
Two oversized captures are no longer treated as unchanged, so the file is listed as oversized instead of vanishing; an oversized side outranks a binary one in the card and the tab alike. Captures read at most maxFileBytes + 1 bytes through a file handle, a missing file resolves through its nearest existing ancestor so a path created through a directory symlink keeps one canonical spelling, and ls-tree matches paths literally. The comparison tab draws at most 5,000 lines. Docs record that comparisons serve whole file text beyond the workspace root, the newline normalization, and the coarse payload bound; the card note's row behavior is current again.
Rows of the changed-files card open a `changes-diff` right-Sidebar tab that compares the file at turn start and turn end. The workspace-changes recorder copies every file a file tool edits into the Session's temporary directory before the first edit and again at turn end, so files git does not cover are compared without git and counted first-to-last; `maxFileBytes` caps copies and snapshot reads, and `diffTimeoutMs` degrades a slow line comparison to whole-file replacement.
A repository whose baseline snapshot failed records nothing instead of
degrading to a file-tool summary; the write tool persists whether it
created or updated, so an unchanged overwrite outside snapshot coverage
no longer counts its content as added; hunks inside nested repositories
are dropped before check-ignore, which refused them with exit 128 and
lost the whole turn; a temporary root inside the work tree keeps the
recorder's own directory out of the snapshots; numstat keeps tabs in
file names; the summary carries line totals over the uncapped list and
the snapshot tree ids; the client's summary store abandons reads that
began before a connection reset. Docs and fixtures follow.
The file tools persist both sides of a hunk without a trailing newline,
so re-diffing an append read the previous last line as rewritten and
reported +2 -1 for +1 -0. Both sides are line-terminated before the
diff, the rule the per-call diff card already applies.
A working directory outside any repository, or a Host without git, no
longer loses the card: the recorder skips the snapshots and builds the
summary from the hunks the file tools persisted, with the working
directory as the workspace, so shell edits are the only gap. The
sidebar e2e scopes its chip clicks to the prose mention now that the
card offers the same open, and the present-svg and permission-policy
scenarios gain the card their write calls earn.
sjw1231