Delete the leftover diff3 parent marker and both copies of the obsolete request-header prompt JSDoc. Keep only the current system/message tokenization contract. This is a local comment-only finding; neither oracle behavior nor fixture content changes.
Evidence: git diff --check is clean; focused normalize.spec.ts scrubSystemPrompts test passes (1 selected, 64 skipped). The earlier header-preservation commit intentionally retained this marker so the findings stay independent.
sessionFixtures already selects the highest generation independently for each parent or child role. Filtering its output to files[0] silently exempted every child from prompt and schema fixed points and prompt-before-request ordering. Merge those assertions into the per-role loop and expose the actual checker for focused negative controls without mocking Vitest registration.
The three selected-child controls reject missing system/message, raw prompt text, and raw tool schemas; each resolved incorrectly with the parent-only filter. A positive mixed-generation case proves retained predecessors remain unselected. Correct the two versionless record-suite child fixtures that the restored enforcement exposes, including the retired-child copy used by recording tests. No released historical generation or broad recording is rewritten.
Evidence: pnpm exec vitest run packages/test-support/session-snapshot/tests/storage-policy.spec.ts packages/test-support/session-snapshot/tests/suite.spec.ts --coverage --coverage.include=packages/test-support/session-snapshot/src/suite.ts: 143 pass, one intentional record-mode skip, suite.ts 100% statements/branches/functions/lines. Update and re-record the session-snapshot README EN/ZH pair.
The loop appends step/start before system/message and the entered user
messages. Capturing nodes at step/start therefore omits inputs already
included in the provider's successful usage, then adds those inputs back
as a positive surface delta. Prompt replacement can also incorrectly add
or subtract the difference from the prior prompt on a completed call.
Snapshot the current priced surface immediately before assistant/message
commits. Keep provider output separate from the durable assistant node so
listener rewrites retain their signed delta. The invariant is zero delta
immediately after an unchanged successful output: provider usage already
includes every admitted prompt input. Later appends/replacements still
produce signed deltas, and low or absent usage keeps heuristic fallback.
Delete stepStart.nodes rather than adding prompt-specific corrections or
another request snapshot: the existing transactional surface fold already
contains the successful request inputs, including replacements made during
same-step retry recovery. Keep turn/step state and all overlap, mismatch,
and late-assistant lifecycle validation. Retry attempts are log-only and
request middleware changes configuration; injected messages remain queued
until admission. No loop, event format, projection, or retry policy changes.
Exercise the real loop with reported usage and initial, growing, shrinking,
and empty prompts; same-step failed attempt plus retry prompt replacement;
request middleware; eager observation and fresh seeded replay. The two
regressions fail before the fix with spurious deltas of +48 and +18 tokens.
Retain existing durable-output rewrite, route repricing, missing/low usage,
transactional failure, and lifecycle tests. Update README EN/ZH and the
existing system-prompt surface-node Agent Note, including pairing records.
Validation (dedicated worktree, no full unit suite):
- pnpm exec vitest run packages/llm/token-meter/tests packages/compaction/compaction-basic/tests/compaction-loop-repro.spec.ts --coverage --coverage.include='packages/llm/token-meter/src/index.ts'
118 passed; exact changed runtime file 100% statements/branches/functions/lines.
- pnpm exec vitest run packages/core/agent-loop/tests/request-reconstruction.spec.ts packages/compaction/compaction-basic/tests/compaction-basic.spec.ts packages/compaction/compaction-basic/tests/loader-composition.spec.ts
118 passed, including retry reconstruction and real Loader composition.
- pnpm run doc-sync: 33 gates passed.
- pnpm run test:docs: 15 gates passed.
- pnpm run lint: passed, 0 warnings/errors.
- git diff --cached --check: passed.
Baseline normalize.ts comment conflict marker is intentionally untouched.
The session already derives the protected system head as a Message, and both adapters accept leading system history. Passing it through a separate SummarizationInput.system string unnecessarily flattens that value and rebuilds the same wire message in the adapter. Prepend the derived head to messages and remove textContent, the separate field, and GenerateOptions.system plumbing from the summarizer.
Keep range selection, shadowed seq accounting, session head protection, routed tools, image references, target policy, and the model-visible compaction instruction unchanged. Empty-content heads still derive to null and contribute no request message, but their surface node remains protected. Update subclass consumers/tests, EN/ZH package and subsystem prose, and the existing system-prompt surface owning note with refreshed pairing records.
Evidence: pnpm exec vitest run packages/compaction/compaction-basic/tests packages/llm/llm-deepseek/tests/serialize.spec.ts packages/llm/llm-pi-ai/tests/context.spec.ts --coverage --coverage.include='packages/compaction/compaction-basic/src/region.ts' --coverage.include='packages/compaction/compaction-basic/src/summarizer.ts' passed 203 tests in 6 files; both changed sources have 100% statements, branches, functions, and lines. Region-to-default-summarizer cases pin exact prefix and tools for nonempty Unicode/multiline, empty, and absent heads. DeepSeek JSON byte equality and pi-ai context equality pin leading-message vs separate-system equivalence on text and image-capable conversion paths.
pnpm run doc-sync passed all 33 gates including doc-typecheck, documentation build, translation pairing and model-experience checks. git diff --check passed. Own dependencies installed with pnpm install --frozen-lockfile. An initial test iteration used a nonexistent ctx.dispose teardown on the in-memory fixture; corrected to its existing fixture lifecycle and reran successfully. No runtime/model behavior, normalizer marker, main worktree, push, or rebase changes.
Remove both unconditional header.system erasures from log normalization and replay comparison encoding, including the unrelated Session-header deletion. The catalog owns released-format migration; comparison must not turn an unexpected field into equality or supply a compatibility shim.
Add direct log/snapshot and catalog-restoration controls that retain unexpected request/header.system and differ from the field-absent fixture. Preserve malformed provenance data as well, closing the owned normalizer coverage gap. Update the two package README pairs with this oracle obligation.
Evidence: pnpm exec vitest run packages/test-support/session-snapshot/tests/normalize.spec.ts packages/test-support/llm-replay/tests/llm-replay.spec.ts --coverage --coverage.include=packages/test-support/session-snapshot/src/normalize.ts --coverage.include=packages/test-support/llm-replay/src/index.ts: 205 tests pass, both files 100% statements/branches/functions/lines. Direct normalizer negative control fails before the fix. Baseline diff3 comment remains untouched for the separate cleanup finding.
Consolidate the representation-change PR and its rebase reconciliations into one baseline. Preserve the exact tree and keep the in-history feature in the dependent PR. Follow-up fixes remain separate.
The failover runbook's Windows switch scope returns to the native Windows jobs:
remove the Python runtime eligibility sentences, the pool Python prerequisite,
and the proposal links. The #3629 self-hosted Python runtime proposal is
retired (deleted as obsolete after its unmeasured-throughput run), and a new
implemented note records that the Windows x64 runtime lane stays on
GitHub-hosted Windows with the evidence and alternatives. Bilingual sidecars
re-recorded.
The failover runbook's Windows switch scope returns to the native Windows jobs:
remove the Python runtime eligibility sentences, the pool Python prerequisite,
and the links to the proposal. The self-hosted Python runtime proposal is
moved to rejected with its unmeasured-throughput verdict, and a new
implemented note records that the Windows x64 runtime lane stays on
GitHub-hosted Windows with the reasons. Bilingual sidecars re-recorded.
Windows x64 runtime builds resolve their hosted matrix.runner unconditionally
again (windows-2025 for pull-request CI). Remove the DSH_CI_FAILOVER_WINDOWS
selector, job-private Python toolchain, self-hosted dependency install and
post-step cleanup, the private setup script, and the routing spec introduced
in #3629. The Windows failover switch again covers only the native Windows
jobs in ci.yml.
Consolidate the existing notes-only PR into one review baseline. Preserve the exact reviewed tree; subsequent corrections remain separate commits with their own rationale and regression evidence.
Reflect the review feedback on the reader list: assistantStreamFirstTokenTime and the has-visible readers stop at the first qualifying member, while lastAssistantStreamChunk, assistantStreamChunks, and joinAssistantStreamText scan the whole stream; assembleAssistantStream feeds a BlockAssembler one joined delta per run. Record-level readers trust the static record type; expandAssistantStream is the validating path.
Session format v2 embeds each attempt's compact stream in
assistant/message and assistant/attempt, but Host and client consumers
still expanded it into per-member TimedStreamChunk arrays and did
per-member work; expandAssistantStream materializes the full array before
find/toReversed/break can answer. Session Stats (the projection phase of
every Session open), the token meter's usage and provider-assembly folds,
the subagent output fold, and the Session Controller image lookup still
paid O(members) allocation and time per settlement.
The Chat and Trajectory definitions were already settled from
message.content on master; the remaining per-member folds stay.
dsh-llm now exports record-level readers (first token, visible content,
visible text, last raw chunk of a type, raw chunks of a type, joined
text, run-aware assembly, per-run first-token/first-visible times) that
scan the compact records once with early exit. Session Stats reads
assistantStreamFirstTokenTime, the token meter reads
lastAssistantStreamChunk(stream, 'usage') and assembles through
assembleAssistantStream, the subagent output fold appends
joinAssistantStreamText, and the Session Controller scans
assistantStreamChunks(stream, 'block-end').
expandAssistantStream is deliberately not memoized: retaining expansions
costs roughly ten times the compact stream for the Session's lifetime.
It remains the validating path at durable boundaries.
Synthetic 200-turn v0 migration benchmark, median of five: first-open
projection 28.0 ms -> 5.4 ms, first-open total 76.9 -> 50.0 ms, peak RSS
137.2 -> 94.9 MB; reopen projection 17.8 -> 5.6 ms; all phase budgets and
the 128 MB heap constraint keep passing.