f0bb0787c9f550c19f7ccfbaf2cfb4db98218db3
2399 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Tom Boucher
|
f0bb0787c9 |
fix(#2640): report truthful state_updated + keep progress frontmatter in sync after phase remove (#2974)
* test(#2640): add regression for state_updated false positive + stale progress Three cases: (1) state_updated reflects actual content change, not just file existence; (2) progress.total_phases resync'd even when the body lacks 'Total Phases:' (the no-op guard was skipping syncStateFrontmatter); (3) state_updated is false when STATE.md doesn't exist. * fix(#2640): report truthful state_updated + force frontmatter resync Two defects in cmdPhaseRemove: 1. state_updated was fs.existsSync(statePath) — trivially true, since the file existed before and readModifyWriteStateMd never deletes it. Now captures the boolean return from readModifyWriteStateMd (changed from void to boolean: true when content was written, false on no-op). 2. progress.* frontmatter stayed stale when the body lacked 'Total Phases:' or 'of N' — readModifyWriteStateMd's no-op guard (#948) skipped syncStateFrontmatter when the body transform was unchanged. Now the transform forces a body diff when a phase was actually removed, so the guard passes and syncStateFrontmatter rebuilds progress.* from the post-deletion disk/ROADMAP state. * fix(#2640): address review — gate forced-diff on targetDir, strengthen assertions Two MAJOR findings from isolated adversarial review: 1. Forced-diff injected a spurious 'Total Phases:' line even when no directory was removed (targetDir === null). Now gated on targetDir !== null. 2. Test #2 asserted 'not 3' instead of '2' — would pass for any wrong count. Now asserts exact value. Test #1 strengthened to assert body Total Phases and frontmatter total_phases both equal 1. * chore(#2640): add changeset fragment * chore(#2640): backfill changeset PR number 2974 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
000a322489 |
fix(#2941): hint at global: prefix when bare skill name matches a global skill (#2973)
* test(#2941): add regression for bare-skill-name global: hint When a bare skill name matches an existing global skill, the skip warning must hint at the global: prefix. When no global skill matches, the original 'Skill not found' warning is unchanged. * fix(#2941): hint at global: prefix when a bare skill name matches a global skill When buildAgentSkillsBlock skips a bare skill name that doesn't exist as a project-relative path, check if it matches an existing global skill. If so, append a hint to the warning: 'a global skill named X exists; use global:X to reference it'. When no global skill matches, the warning is unchanged. getGlobalSkillDir and getGlobalSkillsBase are already imported in this module for the global: branch. The hint is guarded on globalSkillsBase being non-null (runtimes without a skills directory don't support the prefix). * chore(#2941): add changeset fragment * chore(#2941): backfill changeset PR number 2973 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
608be0e7cf |
fix(#2913): distinguish empty cherry-pick from genuine conflict in hotfix create (#2970)
* test(#2913): add regression for hotfix empty-cherry-pick discrimination Two layers: (1) real-git test proving the discrimination logic (check for unmerged paths → skip if empty, abort if conflict) is correct; (2) source-text assertions proving the logic and summary heading are in release.yml. Before the fix: release.yml treats any non-zero cherry-pick exit as a conflict, so an already-applied commit (empty pick) aborts the entire hotfix create. * fix(#2913): distinguish empty cherry-pick from genuine conflict git cherry-pick exits non-zero for BOTH genuine conflicts AND empty picks (the change is already present by content). The hotfix create job treated every non-zero exit as a conflict, so an already-applied commit (notably the structural 'chore: sync next package version' that follows every release finalize) aborted the entire run. Now the error handler checks for unmerged paths (git diff --diff-filter=U): - No unmerged paths → already applied by content → skip, record, continue. - Unmerged paths present → genuine conflict → existing abort/push/exit-1 behavior and operator guidance, unchanged. The job summary now has a separate 'Skipped (already applied by content)' heading, distinct from 'Skipped (feat/refactor/etc)'. * fix(#2913): address review — post-skip continuation test + conflict observability Two findings from isolated adversarial review: 1. MINOR (test gap): no test proved the sequencer is clean after --skip, so a regression switching --skip to --quit would pass green. Added a second cherry-pick after the skip asserting it succeeds. 2. MINOR (observability): SKIPPED_EMPTY was dropped on the conflict-exit path — already-applied commits before a genuine conflict were silently lost from the summary. Now the conflict summary emits them under a dedicated heading. * chore(#2913): add changeset fragment * chore(#2913): backfill changeset PR number 2970 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
d3305fc3a5 |
fix(#2858): exclude gen-emitted-baseline.cjs from npm tarball + class-extinction guard (#2968)
* test(#2858): add class-extinction guard for shipped-script require boundary Every shipped scripts/**/*.cjs must be require-able using only shipped paths. The guard resolves the tarball file list via npm pack --dry-run --json (not a hardcoded list) and statically checks each require() call against the shipped set. A script requiring ../tests/** (which does not ship) is a violation. * fix(#2858): exclude gen-emitted-baseline.cjs from the npm tarball scripts/gen-emitted-baseline.cjs is repo-only CI tooling (CI workflows + test fixtures spawn it from a checkout). It requires three modules from tests/, which does not ship — so in a published install it is MODULE_NOT_FOUND at load time. Add a targeted files[] negation (!scripts/gen-emitted-baseline.cjs) so the script stays in the repo for CI use but does not ship. Other scripts that ship and are required by bin/install.js (build-hooks.js, fix-slash-commands.cjs, gen-capability-registry.cjs) are unaffected. The class-extinction guard test in tests/packaging-shipped-scripts-require-only-shipped.test.cjs ensures no shipped script can require outside the shipped tree going forward. * test(#2858): widen guard to .js + strip block comments (review fixes) Two findings from isolated adversarial review: 1. MAJOR: the guard only checked .cjs files, but scripts/build-hooks.js ships and is required by bin/install.js — a .js file with a broken require would bypass the guard. Widened filter to /\.(cjs|js)$/. 2. MODERATE: the static parser could false-positive on require() calls inside /* */ block comments or inline // comments. Now strips both before matching. * chore(#2858): add changeset fragment * chore(#2858): backfill changeset PR number 2968 * fix(#2858): add issue ref to allow-test-rule exemption (ADR-456) CI lint-tests caught: the allow-test-rule comment needs a 'see #NNN' ref per ADR-456. Added '(see #2858)' to the integration-test-input exemption. --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
1db7dcd9bf |
fix(#2849): strip trailing hyphen after 60-char slug truncation (#2967)
* test(#2849): add failing regression for trailing-hyphen-after-truncation The strip ran before .substring(0, 60), so a cut landing on a separator produced a slug ending in '-'. Four cases: the exact issue repro (59 a's + space + tail), a boundary landing before a separator, leading-hyphen survival, and a long-Cyrillic transliteration+truncation case. * fix(#2849): strip trailing hyphen after 60-char truncation generateSlugInternal ran the ^-+|-+$ hyphen strip BEFORE .substring(0, 60), so a title whose 60-character cut landed on a separator yielded a slug ending in '-' — the very thing the strip step exists to prevent. Reorder so the strip runs after truncation. Truncation cannot introduce a leading hyphen, so the full ^-+|-+$ pass last is equivalent for leading hyphens and fixes the trailing-hyphen-after-truncation case. Latin-script output for titles ≤ 60 chars is byte-identical; only titles whose truncation boundary lands on a separator change (from broken to clean). * test(#2849): add all-separator collapses-to-empty boundary case Surfaced by isolated adversarial review: pin the contract that input which is entirely separators ('!!!', '!'.repeat(70)) reduces to '' — not null, not a stray hyphen — both short and past the 60-char truncation. * chore(#2849): add changeset fragment pr:0 placeholder; will backfill the real PR number after the PR exists. * chore(#2849): backfill changeset PR number 2967 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
0bb7525a62 |
fix(#2943): rename get-library-docs -> query-docs; correct the ctx7 fallback rationale (#2963)
* test(#2943): parity guard against the nonexistent get-library-docs tool Second context7 naming drift after #2017 (which guarded the plugin-marketplace PREFIX). #2017's guard only checks tools: frontmatter lines, not prose bodies — which is where the broken tool NAME (get-library-docs) lived. The context7 MCP server registers only resolve-library-id and query-docs; get-library-docs is a stale copy from upstream's own README. Scans the shipped prose surface (agents/, gsd-core/references|workflows/, commands/gsd/, skills/) and fails if any artifact instructs an agent to call mcp__context7__get-library-docs. Excludes tests/ (a fixture may use the name as a negative input) and CHANGELOG/RELEASE-NOTES-LEGACY (history). Fails-first: 4 offenders today (gsd-executor.md:29, research-documentation-lookup.md:5, discovery-phase.md:68 & :104). * fix(#2943): rename get-library-docs to query-docs and correct the ctx7 fallback rationale The context7 MCP server registers only resolve-library-id and query-docs (verified against upstream packages/mcp/src/index.ts); get-library-docs is a stale name copied from upstream's own README. Four shipped prose sites instructed agents to call a tool the server does not register, so every research path that loaded the canonical reference either errored, fell through to the ctx7 CLI branch, or fabricated a result. - research-documentation-lookup.md, gsd-executor.md, discovery-phase.md (x2): get-library-docs -> query-docs, params context7CompatibleLibraryId/topic -> libraryId/query (the registered contract). - Same files' ctx7 CLI fallback rationale: the cited cause (anthropics/claude-code#13898 'strips MCP tools from agents with a tools: frontmatter restriction') was wrong on two counts — #13898 is closed and was never about tools: frontmatter. Rewritten to describe the real mechanism (custom subagents cannot see project-scoped .mcp.json; they only inherit user-scoped ~/.claude/mcp.json). The fallback itself is kept. - discovery-phase.md 'mode: code/info' dropped — query-docs takes libraryId + query only; the code-vs-concepts intent is now expressed via the query text. resolve-library-id is unchanged (still registered upstream). CHANGELOG and RELEASE-NOTES-LEGACY citations are historical record, left as-is. * chore(#2943): add changeset fragment (pr:0 placeholder) * test(#2943): widen parity-guard scan surface to docs/ (isolated-review finding) The isolated adversarial review flagged that SCAN_DIRS omitted docs/, which ships docs/AGENTS.md — agent-consumed prose carrying 8 mcp__context7__* refs. No false negative today (it uses only the wildcard), but a future banned-name addition there would slip through, recreating the exact drift this guard exists to prevent. Add docs/ to the scan surface, with an EXCLUDED_FILES set for historical record (docs/RELEASE-NOTES-LEGACY.md, CHANGELOG.md) that must not be rewritten to satisfy the guard. * fix(#2943): update shifted PROSE_ALLOWLIST line + acknowledge gsd-executor.md growth The gsd-test gate caught two real consequences of the rationale rewrite in agents/gsd-executor.md (the +2-line corrected mechanism description shifted line numbers below it): 1. tests/no-bare-gsd-tools-command-position.test.cjs: the legitimate 'gsd-tools query commit' descriptive mention moved from line 791 -> 793. Update the PROSE_ALLOWLIST entry to the new line (the mention is unchanged, just relocated by my edit above it). Without this the gate reports both a stale allowlist entry (791) and a new offender (793) for the same mention. 2. tests/emitted-drift-acks/2943-context7-tool-name.json: gsd-executor.md grew 95 bytes (the accurate mechanism rationale is longer than the wrong one-line #13898 attribution it replaces). Acknowledge the growth with the reason. Both are mandated by the gate, not optional. The rename itself (get-library-docs -> query-docs) is byte-neutral-ish; only the rationale rewrite grew the file. * chore(#2943): backfill changeset PR number 2963 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
73418c516f |
fix(#2956): scope Phase extraction to ## Current Position (3rd gen of #2444/#2567) (#2961)
* test(#2956): fail-first regressions for Phase scoped to ## Current Position Third generation of #2444 / #2567. Stopped At / Paused At were scoped to ## Session; Phase (canonically in ## Current Position per templates/state.md) was left unscoped, so a historical Phase: / **Phase:** line in an archive section overwrites current_phase on every write. Since current_phase is routing input for gsd-progress / --next, the rewind routes work to the wrong phase. Six failing-first regressions + one round-trip: - shape B: bold **Phase:** 19 archive BELOW the section - shape C: plain archive Phase: 19 ABOVE the section - bootstrap h3 ### Current Position variant - CRLF variant - Phase token in decisions prose (over-broad-fix guard) - Paused At read-path parity with the write seam (## Session) - write-then-read round trip stays at 22 (read/write agreement) Folded into tests/state.test.cjs (lint:regression-test-names bans a new tests/bug-NNNN-*.test.cjs file). * fix(#2956): scope Phase extraction to ## Current Position at both seams Third generation of #2444 / #2567. Stopped At / Paused At were scoped to ## Session by those fixes; Phase (canonically in ## Current Position per templates/state.md) was left unscoped, so a historical Phase: / **Phase:** line in an archive section silently overwrote current_phase on every write. Because current_phase is routing input for gsd-progress / --next, the rewind routes work to the wrong phase. Fix mirrors the proven #2444 seam exactly: - new matchCurrentPositionSection helper (collectSection-based, CRLF-tolerant, level-flexible for the bootstrap ### Current Position h3 variant), sited next to matchSessionSection. - read path (cmdStateSnapshot): extract Phase from matchCurrentPositionSection ?? body. Also scope Paused At to matchSessionSection ?? body so the read seam agrees with the write seam (which already scoped Paused At to ## Session). - write path (buildStateFrontmatter): extract Phase from matchCurrentPositionSection ?? bodyContent. stateExtractField itself is untouched (its bold/plain precedence is load-bearing for other fields — the #3265 test depends on it), and preferNewerLastActivity is untouched (Last Activity has no canonical section; its date-direction guard is deliberate). Fall back to full-body when no ## Current Position section exists so files without the heading keep current behaviour. * chore(#2956): add changeset fragment (pr:0 placeholder, backfill after PR) * test(#2956): make round-trip test actually trigger the write-path resync The write-then-read round-trip test used 'state update Status "Executing"' on a fixture with no Status field, so the update was a no-op (updated:false) and no frontmatter resync ran through buildStateFrontmatter — the assertion on the written frontmatter then failed not because the fix is wrong, but because no write happened. Add a **Status:** field so the update performs a real field update (updated:true) and forces the resync. Verified locally: pre-fix this writes current_phase:19 (the archive value); post-fix it writes 22. The code fix is correct (5 of 7 RED tests passed; the 2 failures were this defective test). This is the 'fix the bad test' half of the TDD-loop rule. * chore(#2956): backfill changeset PR number 2961 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
9ac0dfad58 |
chore(#2929): generalize prompt-budget into the shared context-composer seam (#2958)
* test(#2929): capture prompt-budget parity corpus pre-refactor Phase 2 of epic #1671 generalizes prompt-budget's trim ladder into a shared context-composer seam. Its success condition is that review-prompt output does not change, and the only authority on "did not change" is the behavior that shipped before the refactor. Capture that behavior now, while it is still the live implementation. 47 characterization cases, every `expected` value computed by executing the current implementation rather than hand-authored — the independence CONTRIBUTING.md "Fixture provenance (#2371)" asks for. A corpus is only worth what it can detect, so this one was validated by mutation rather than assumed. Five deliberate defects were injected and each must be caught by at least one case: - the note reserve deducted unconditionally instead of only under pressure - the pressure test relaxed from `>` to `>=` - a no-op head-shrink still setting the shrunk flag - the per-plan floor dropped from the proportional share - drop order reversed Two of those exposed real holes in the first cut of this corpus, and the cases that close them exist because of it: - `>=` was caught by NOTHING. At exact cap the only trimmable fragment was a floored plan group, and the 1024-char floor absorbed the entire trim, so the mutation was byte-invisible. A3b/A3c put a droppable at exactly the cap, which makes the strict inequality observable as context kept vs omitted. - No case reached proportional-truncate at all — B6 and B7 both hard-failed the min-set pre-check first, leaving planTruncationPct at 0 across every case and the floor semantics entirely unexercised. Rebudgeted to 700 and 1100 so the min-set fits and the truncate step is actually reached; they now record 40.20% and 48.80%. The A4/A10 families sweep the pressure boundary from both sides, which is where this function has regressed before: CONTEXT.md's LEARNING.prompt-budget.boundary-gap records PR #3708 shipping two regressions that only fired when the baseline sat inside the NOTE_RESERVE_TOKENS band, because the suite paired a trivially-fitting budget with a trivially-overflowing one and never sampled between them. A4 pins that nothing is trimmed from the cap down to 81 tokens under it; A10 pins that pressure fires at +1. Together with A3b/A3c they satisfy row (d) of RULESET.TESTS.boundary-coverage.fixtures. Two facts the corpus establishes that the design notes had wrong: - "" and null sections are NOT distinguished. applyBudget uses truthy checks throughout, so an empty-string section is treated as absent: not rendered, not dropped, never recorded in `omitted`. B13b pins this while the ladder is actively trimming, where only the non-empty `research` is dropped. - Sizing matters. B12/B13 were first written at a budget where both hard-failed the min-set check and returned "", so comparing them compared two empty strings and proved nothing. Committed as its own commit, ahead of the refactor, and regenerated against the pre-refactor implementation, so the oracle is demonstrably independent of the change it will adjudicate. Refs #2929 * refactor(#2929): extract the context-composer seam from prompt-budget Epic #1671 needs prompt-budget's budget-trimming logic for a second consumer — per-runtime artifact emission — but it is walled inside the cross-AI review pipeline. Lift it into a shared seam so later phases can call it, without changing what the review pipeline emits. ADR-1671 specifies the composer as "priority + binary-search cutoff to a per-runtime budget". Read against the code it generalizes, that contract cannot express the thing being generalized. applyBudget is not a cutoff: it is a fixed five-step ladder in which each section carries its own shrink strategy, and only three of its eight sections are ever dropped. PROJECT.md is head-shrunk to N lines; plans are proportionally tail-truncated with a per-plan 1024-byte floor; instructions and roadmap are never touched at all. A cutoff composer sorts by priority and discards the tail — it has no way to say "shrink this one", "truncate that one but never below 1 KB each", or "these three are the only droppables, in this order". Building to the literal contract and routing prompt-budget through it would have silently changed review-prompt output, which is the one outcome this phase forbids. So shrink strategies are the core abstraction here, and cutoff becomes one strategy among them — the right one for per-runtime emission in Phases 3-4, not for this ladder. That is an elaboration of the ADR's intent, not a departure from it, and ADR-1671 is updated to say so. Three decisions worth stating: - The composer DECIDES; the caller RENDERS. composeWithinBudget returns a plan of surviving fragments and never a string. assemblePrompt's rendering is prompt-shaped (`## Roadmap`, `### <file>`, the note in position two), and owning it in the composer would force emission to adopt prompt-shaped rendering. The split is what lets one seam serve both consumers. - The budget unit is INJECTED via `measure(text)`. prompt-budget passes its chars/4 estimator; emission will pass a byte counter, which ADR-1671 requires for emission caps. The existing code converts a token budget to a character budget with a hardcoded `* 4`; that assumption is now an explicit `charsPerUnit` inverse, which is precisely what a byte unit needs in order to reuse this. - The entry point is `composeWithinBudget`, not `applyBudget`. That name already exists twice — src/prompt-budget.cts and src/graphify.cts, the latter being an unrelated graph-edge budget. A third would make every symbol search in this repo ambiguous, and it already misresolves: preflight and impact queries for "applyBudget" return graphify's. Behavior is unchanged and proven so: all 47 characterization cases reproduce byte-identically, and the corpus is mutation-validated rather than merely green (see the preceding commit). prompt-budget.cts drops from 436 to 343 lines and from eighteen mutable accumulators to two, both inside a helper copied verbatim. estimateTokens deliberately stays in prompt-budget and keeps its exact math: src/phase-estimation.cts re-exports it as measureTokens, and CONTEXT.md pins plan estimates and recorded actuals to that same scale, so moving or changing it would silently break the calibration loop. Refs #2929 * docs(#2929): document the context-composer seam and amend ADR-1671 Adds the INVENTORY row, the CONTEXT.md glossary entry (a PR gate for new domain modules), and a mutation-matrix entry for the new module. The ADR amendment is the substantive part. ADR-1671 specified the composer as "priority + binary-search cutoff to a per-runtime budget". Implementing Phase 2 established that a cutoff alone cannot express the function the platform generalizes, so the ADR now records shrink strategies as the core abstraction with cutoff as one strategy among them, reserved for per-runtime emission in Phases 3-4. Recording it in the ADR matters because Phases 3-6 are planned against that contract and would otherwise be planned against a mechanism that does not work. The mutation-matrix entry is not bookkeeping. Stryker scores per module against a named .cjs, so relocating the ladder out of prompt-budget.cjs would leave the extracted code unmeasured while prompt-budget's own score floated free of the logic it used to cover. context-composer gets its own entry at the same floor. Refs #2929 * test(#2929): pin the effectiveBudget rounding mode in the parity corpus An isolated correctness review found a real blind spot: mutating `Math.floor` to `Math.round` in the effectiveBudget calculation failed ZERO of the 47 corpus cases. Every (budget, safetyMarginPct) pair in the generator happened to produce a whole number, so floor, round and ceil all agreed and the rounding mode was entirely unpinned by a corpus whose whole job is to pin observable behavior. Three cases fix that by straddling the .5 boundary: A11 95 * 0.90 = 85.5 floor 85, round 86 -> the two disagree A12 97 * 0.90 = 87.3 floor and round agree; ceil (88) does not A13 93 * 0.85 = 79.05 same guard at a non-multiple-of-10 margin, so the margin arithmetic is exercised and not just the budget A11 alone catches the round mutation; all three catch ceil. Regenerated against the pre-refactor implementation (`git show 9557f8552:src/prompt-budget.cts`), so the expanded corpus keeps the independence property the original capture had. The corpus is now mutation-validated against seven injected defects, every one caught: unconditional note reserve, `>` relaxed to `>=`, no-op head-shrink setting its flag, the truncate floor ignored, drop order reversed, and both rounding-mode changes. Refs #2929 * feat(#2929): flexReserve floors and the byte-stable isolate prefix Two of issue #2929's "Done when" items were unimplemented rather than deferred, and an isolated review flagged them alongside my own audit. Both are part of ADR-1671's composer contract, so shipping the seam without them would have left Phases 3-4 building against a contract that does not exist yet. flexReserve is a per-fragment floor in measure units that every strategy must respect, which is what makes it different from the pre-existing floorChars: that one is a chars-denominated detail of proportional-truncate alone and is retained unchanged. A floored fragment is never dropped, is never head-shrunk below its floor, and raises its own proportional cap. A fragment already smaller than its floor is untouchable outright. Metadata gains `floored`, listing the ids whose floor actually prevented a trim — a guarantee no caller can observe is a guarantee no test can hold you to. isolate marks the byte-stable canonical prefix the ADR calls for: never trimmed, never dropped, but still counted, because a prefix excluded from accounting would silently under-count real context. Metadata gains `isolatePrefix` so a caller can hash or assert on the exact bytes. Declaring an isolate fragment after a non-isolate one throws: a prefix that is not at the front is not a prefix, and accepting it would make the cross-runtime stability claim meaningless. Adds tests/context-composer.test.cjs for the exact new semantics and tests/context-composer.property.test.cjs for the five invariants, including the budget-monotonicity property the issue names explicitly. Both are registered in the mutation matrix, since coverage does not migrate with relocated code. prompt-budget uses neither feature, and its output is unchanged: all 50 corpus cases still reproduce byte-identically. Refs #2929 * chore(#2929): allowlist the prompt-budget parity suite The parity corpus needs its own test file and that makes prompt-budget a three-file module against a limit of two. The lint offers consolidation or an allowlist entry with justification; the entry is the right call here. Consolidation would mean folding the characterization suite into prompt-budget.test.cjs, which is the one thing that should not happen to it. The parity suite is a distinct concern with a distinct lifecycle: it is generated rather than hand-written, it is named by scripts/mutation-matrix.cjs as its own scoring target, and its failure means something categorically different from a unit-test failure — not "this behavior is wrong" but "observable output moved". Burying it inside a general unit file would obscure exactly that signal. The allowlist is an identity ratchet, so this entry pins today's three exact filenames: adding a fourth still fails, and dropping back to two requires removing the entry. Refs #2929 * fix(#2929): register the new module with two gates it was missing The remote matrix caught three defects that no local check could, because the local runner is blocked in this repo and these suites had therefore never executed. Eight failures, identical on node22 and node24, so nothing environment-shaped. Two are the new-module ripple. A net-new src/*.cts lands in six places and this change had reached four of them — .gitignore, INVENTORY, the manifest, and the CONTEXT.md glossary — while missing the ESLint ignore list (tsc OUTPUTS must not be linted; repo-invariants asserts linted-xor-ignored) and the mutation ratchet baseline (a deliberate review-visible mirror of the matrix floors, which every COVERED module must carry). Both are now registered, the ratchet at the same floor of 66 the matrix declares. The third was a test asserting an outcome it had made impossible. It set budget:1 alongside a 400-char required fragment, so the group budget came out at -99 and the proportional-truncate step was skipped entirely — the deliberate "non-positive group budget is skipped, never clamped" rule inherited from the original ladder. Nothing was trimmed, and the test then asserted a truncation. Rebudgeted so the step actually runs, with the arithmetic written out in a comment so the next reader does not have to re-derive why 120 rather than 80. Fixing that surfaced a genuine bug in the composer. `floored` is documented as recording fragments whose flexReserve prevented a trim that would otherwise have happened, but the push sat in the else-branch of "content did not change", so it only fired when nothing was trimmed at all. A fragment truncated to a reserve-raised cap has also had a trim prevented — 40 characters' worth in the test above — and was silently absent from the field that exists to make the guarantee observable. The condition was already right; it was in the wrong branch. Now recorded on both paths: a drop prevented outright, and a truncation capped higher than the share alone would have allowed. Parity is unaffected — prompt-budget never sets flexReserve, so the branch is unreachable from every corpus path, and all 50 cases still match. Refs #2929 * chore(#2929): backfill changeset PR number (#2958) * chore(#2929): correct the corpus case count in the changeset fragment --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
f6257f3745 |
ci(#2952): shard the full test lane instead of widening its cap (#2960)
* ci(#2952): budget CI job timeouts by headroom over measured cost `origin/next` was red. The only failing check was `Required tests`, and its sole cause was `test (ubuntu-latest, 24)` reported `cancelled` — GitHub's conclusion for a job that exceeds its `timeout-minutes`, not a button press. That job's `ubuntu-latest / 24` entry is the only `scope: full` matrix entry: it runs the whole unit suite under c8 coverage, then the scripts/ coverage floor, integration, security, install and slow, serially on one runner. On next@5a0a9f097 it ran 15m16s against `timeout-minutes: 15` and was axed 23s into `npm run test:slow`. Projected to a completed slow step (27s on the last green run) the lane costs ~15m20s. Confirmed hypothesis: the budget, not the suite. The lane had been riding the ceiling all day — 12m03s, 11m34s, 11m50s, 14m25s, 14m51s — and crossed on three of the last four full-lane runs ( |
||
|
Daniel Einspanjer
|
f0ff23635e |
fix(#2602): discover project-local Codex agents (#2623)
* fix(#2602): discover project-local Codex agents - Select an existing local Codex agents directory before global fallback - Prove init reports the canonical local installation through compiled CJS * test(#2602): lock Codex agent precedence - Cover override, local authority, global fallback, and runtime compatibility - Exercise installed state through the compiled resolver * fix(#2602): resolve local Codex agent skills - Pass the canonical project root to the non-Claude persona fallback - Cover nested-Codex fallback and Claude compatibility through the CLI * test(#2602): cover local Codex validation status - Assert emitted validate and health commands use the project-local install - Preserve empty local-directory authority beside complete global agents * fix(#2602): align validation with local Codex discovery - Pass the resolved runtime and project root to health W010 - Resolve the validate-agents runtime before checking installation status * test(#2602): cover local Codex docs status - Assert docs-init reports an authoritative empty local install as unhealthy * fix(#2602): align docs with local Codex discovery - Pass the resolved runtime and canonical project root to the shared agent checker * fix(#2602): honor agent-skills runtime override - Resolve agent-skills fallback runtime through the canonical project resolver - Cover conflicting config and GSD_RUNTIME values through the emitted CLI * fix(#2602): ignore non-directory local agents paths - Treat only a local Codex agents directory as authoritative - Cover regular-file fallback through the emitted install checker * chore(#2602): add changelog fragment - record the user-visible local Codex agent discovery fix for PR #2623 * fix(#2602): align local agent discovery with runtime policy - Resolve Codex's local config directory through the canonical runtime policy - Use test-managed cleanup for local-agent discovery coverage * fix(#2602): discover local agents across runtimes - Prefer manifest-backed project-local installs for non-Claude runtimes - Respect runtime-specific local install roots and preserve global fallback behavior - Cover native, partial, cross-runtime, and project-root local discovery * fix(#2602): preserve agent discovery fallback - Fall back globally when local-install probes fail - Document and test symlink rejection - Align the changeset with repository format * fix(#2602): reuse local directory policy - Resolve runtimes without local config through the canonical sentinel - Document the manifest gate and refresh the context index --------- Co-authored-by: Daniel E. <daniel.e@teachingstrategies.com> Co-authored-by: Rezolv <dave@sienkowski.com> |
||
|
JusticeWay
|
7b204ad2ac |
enhance(#2530): extend UAT checkpoint frame language pack (9 more languages) (#2564)
* feat: extend UAT checkpoint frame language pack (9 more languages) response_language is a free-form config value, but CHECKPOINT_FRAMES only covered 9 languages — any other configured language silently fell back to the English frame. Add Dutch, Polish, Russian, Ukrainian, Turkish, Hindi, Arabic, Vietnamese, and Indonesian frames plus their aliases, with a regression test asserting each resolves instead of falling back. Follow-up to #2402 (PR #2457). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore: add changeset for #2527 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs(#2530): list UAT checkpoint frame languages in CONFIGURATION.md Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(#2530): point changeset fragment at PR #2557 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(#2530): address Unicode language-pack review * fix: address checkpoint language review * fix: count spacing combining marks in checkpoint width * test: verify checkpoint aliases structurally * fix: isolate RTL checkpoint frames * fix: isolate RTL checkpoint frames correctly * test(#2530): assert checkpoint aliases neither collide nor go unreachable Review Minor #1. A duplicate alias key was invisible to the existing catalog tests: the runtime object is well-formed after JS collapses the literal, the self-alias assertion still holds, and the losing language just stops resolving. tsc catches the byte-equal case (TS1117), but not the two that survive compilation — an alias whose NFC-lowercase form already belongs to another language, and an alias not in lookup form at all, which resolveCheckpointFrame() can never produce. The check reads the source literal rather than the object, since the object no longer records what was written. Both assertions are independently load-bearing: an NFD twin of an existing alias trips the collision check, an uppercase alias trips the unreachability check. Review Minor #2: changeset retyped Changed -> Added. Nine wholly new supported response_language values are an addition under Keep a Changelog, not a modification of existing behavior. * test(#2530): check alias collisions on the catalog, not its source --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> Co-authored-by: Tom Boucher <trekkie@nomorestars.com> Co-authored-by: Rezolv <dave@sienkowski.com> |
||
|
Tom Boucher
|
f092c6da85 |
fix(#2649): diagnose-issues + execute-plan run worktree.base-check before worktree dispatch (#2955)
* test(#2649): failing-first — diagnose-issues + execute-plan must run base-check before worktree dispatch * fix(#2649): diagnose-issues + execute-plan run worktree.base-check before dispatch diagnose-issues.md spawn_agents and execute-plan.md Pattern A spawned worktree-isolated subagents (gsd-debugger / gsd-executor) without the pre-dispatch worktree.base-check gate that execute-phase (#683/#1369) and quick (#1941) already run. Claude Code's isolation="worktree" forks from origin/HEAD, not live local HEAD; without the gate, the documented GSD steady state (commit every step locally, push only on request) hits the verify-only worktree_branch_check guard's exit-42 halt mid-investigation with no auto-degrade. Mirror the quick.md #1941 pattern: before dispatch, run `gsd_run query worktree.base-check --pick shouldDegrade`; if true, print its message + a #2649 warning to stderr and set USE_WORKTREES=false (sequential main-tree dispatch). The verify-only guard stays as a backstop in both cases. Per the triage and #2649 acceptance criterion 5, execute-plan.md's Pattern A (identified as a second site with the identical gap) is fixed in the SAME change — same bug class, same one-line gate, two workflow files — rather than filed as a separate follow-up. * fix(#2649): ack the diagnose-issues + execute-plan growth (per-PR fragment) The two workflow files grew vs next (diagnose-issues.md +1381, execute-plan.md +905) adding the #2649 base-check gate. emitted-attribution requires an ack; this is a per-PR fragment under tests/emitted-drift-acks/ (#2914 mechanism, replacing the legacy shared emitted-drift-ack.json). * test(#2649): tighten base-check ordering assertion + guard backstop survival Address code-review minors: - the ordering assertion was a loose disjunction that passed even if the base-check moved AFTER the dispatch; tighten to assert base-check < Agent() (the real invariant). - add a test that the verify-only <worktree_branch_check> backstop remains embedded in the Agent() prompt (acceptance criterion 4 — the base-check is a pre-dispatch degrade, the guard is a post-fork fail-closed backstop; both layers must survive). * changeset(#2649): diagnose-issues + execute-plan auto-degrade on stale worktree base * changeset(#2649): backfill PR number 2955 --------- Co-authored-by: sim <sim@users.noreply.github.com> |
||
|
Tom Boucher
|
388837219d |
fix(#2648): phase.complete refuses when non-retired plans lack summaries (fail-closed coverage gate) (#2953)
* test(#2648): failing-first — phase complete must refuse when plans lack summaries Adds findUnsummarizedPlans to core-utils (mirrors countMatchedSummaries but returns the unmatched plan files) and a PHASE_PLAN_COVERAGE_INCOMPLETE error reason, plus a 3-case regression block in phase.test.cjs. The gate itself is NOT yet wired into cmdPhaseComplete (reverted for the RED run), so the 'blocks completion' and 'superseded does not block' cases must FAIL (the pre-fix code completes silently). * fix(#2648): phase.complete refuses when non-retired plans lack summaries cmdPhaseComplete gated only on a single *-VERIFICATION.md status, so a phase could close complete while an arbitrary number of its plans had no completion record (confirmed incident: 6/30 plans unexecuted incl. the phase's entire final UI scope, every signal green). Add a fail-closed plan-coverage gate that refuses completion when any plan lacks a matching *-SUMMARY.md, naming the missing plans, UNLESS the plan is retired via machine-readable status: superseded frontmatter (#2349) — closing the Goodhart hole (delete a SUMMARY to raise the %) without regressing the lock/recovery pattern. Uses scanPhasePlans (superseded-AWARE) + new findUnsummarizedPlans helper so the gate, the count, and the named list can never disagree. Evaluated before the verification-gate transaction so a refusal fails fast without mutating ROADMAP/STATE. milestone.complete's parallel gap is explicitly out of scope (separate seam, separate PR). * fix(#2648): test fixtures — give #1752 phase plans summaries + STATE.md in coverage fixture The plan-coverage gate (#2648) correctly blocks phase completion when a plan lacks a SUMMARY. Two test fixtures needed updating to reflect the new contract: - #1752 (total_phases-decrement cascade): its 8 phase dirs each had a PLAN.md with no SUMMARY. The test's concern is the total_phases cascade, not plan coverage, so add a matching SUMMARY to each to keep the phase fully-covered and isolate the #1752 behavior. - the #2648 coverage-gate fixture: write STATE.md (createTempProject scaffolds .planning/phases but not STATE.md) so the 'ROADMAP/STATE unchanged on refusal' assertions have a file to read. * fix(#2648): security — fail closed on unreadable plan dir + sanitize msg + surface superseded Address the security-review blocker (B1) and hardening (M1/m2): - B1 (blocker): the gate failed OPEN when scanPhasePlans could not read the phase dir (it swallows readdirSync errors → empty plan set → gate sees zero unsummarized plans → passes). A coverage gate that passes when it cannot read the plans re-opens the #2648 hole under any I/O failure. Now readdirSync the dir explicitly and fail closed (PHASE_PLAN_COVERAGE_INCOMPLETE) on a throw; a readable empty dir still passes (legitimately complete empty phase). - m2: sanitize plan filenames (strip C0 controls / DEL) before interpolating into the error message — they come raw from readdirSync and could spoof the terminal in plain-error mode. - M1: surface the count of plans excluded as status: superseded so a reviewer can audit which work was declared retired (the marker is a committable, review-time-trusted bypass; keep it visible). - Add a 4th regression case: unreadable plan dir (ENOTDIR via a file, not chmod 0o000 which root bypasses) must fail closed. * test(#2648): drop unreachable B1 case — no root-safe unreadable-dir repro The B1 fail-closed-on-unreadable-dir defense stays in src/phase.cts (cheap + correct), but it cannot be unit-tested cross-platform: any condition that makes the phase dir unreadable to the gate's readdirSync ALSO fails findPhaseInternal upstream ('Phase N not found') before the gate runs, and chmod 0o000 is forbidden (root bypasses it in root CI). Document the gap in the test file; remove the case that asserted a reason the upstream error pre-empts. * style(#2648): drop unnecessary type assertions flagged by lint:ci scanPhasePlans returns typed string[] arrays, so the `as string[]` casts on coverageScan.planFiles/summaryFiles were redundant (@typescript-eslint/ no-unnecessary-type-assertion). Compute supersededCount from typed lengths; only the phaseInfo['plans'] cast remains (it is genuinely unknown). * changeset(#2648): phase.complete refuses when plans lack summaries * changeset(#2648): backfill PR number 2953 --------- Co-authored-by: sim <sim@users.noreply.github.com> |
||
|
Tom Boucher
|
5a0a9f0972 |
fix(#2944): remove the catastrophic-backtracking regex from the ADR-1671 example (#2950)
* fix(#2944): remove the catastrophic-backtracking regex from the example The non-shipping Option-E reference example carried its own copy of the predicate-id regex, which nested a dot-containing character class inside a dot-prefixed repeat. A run of N consecutive dots therefore had exponentially many partitions. Measured on next before this change: 30 dots 54ms, 35 66ms, 40 807ms — so roughly 55-60 dots hangs for hours. Not exploitable where it sits: the example is outside tsconfig.build.json, outside the npm package files list, outside the installer and outside tests, so no build step or CI job parses anything with it. Fixed because the entire point of a reference example is that people copy it forward, and ADR-1671 presents this one as the pattern for the platform. Ports the linear per-segment validation that #2928 gave the production module, so the two copies agree: both parse the real CONTEXT.md to 415 predicates across 20 classes with 0 duplicates. Doubled-dot ids are now rejected here too, matching production, and the grammar comment records it. Also refreshes the example's committed index, which #2928 made stale when it removed the duplicate predicate from CONTEXT.md. Closes #2944 * test(#2944): guard predicate-index sync and example/production parity Two regression tests for the two defects in this PR. Index sync: asserts the committed docs/CONTEXT-INDEX.json equals a fresh parse of CONTEXT.md, naming any diverging predicate ids. The merge race that reddened next was invisible to both PRs involved and only surfaced on the next PR to run lint:ci; this puts the same check inside the suite, which runs on every PR, and a mutation test proves the assertion is not vacuous. Example/production parity: asserts both copies of the parser report the same count, classes and duplicates for the real CONTEXT.md, and agree verdict-for- verdict over a table of id shapes. The divergence WAS the bug — production went linear-time while the example kept the backtracking regex, with nothing asserting they agreed. Also pins the example rejecting a 60-dot id, with the clean rejection as the binding assertion and wall-clock only as a smoke check. Notes a real tension rather than hiding it: ADR-1671 says the example sits outside tests/, and this imports it. The ADR's intent is that the example is not compiled, packaged or installed — not that it may silently rot. A parity guard does not ship it. The file states this so a reviewer can object. * fix(#2944): address both isolated review passes Two independent reviewers (correctness and security axes, neither the author). Security found nothing — it measured linearity to 100k chars across dots, hyphens, underscores and mixed classes, and showed prototype pollution is structurally unreachable because the first-segment pattern forbids lowercase and underscore-leading ids. The correctness pass found three blockers, all real. Blocker: the parity test violated ADR-1671 verbatim. The ADR lists FOUR exclusions for the reference example, the fourth being the CI test suite, and the test imported it from tests/ while its own justification comment cited only three -- constructing a rationale around the exclusion it broke. Moved to scripts/lint-example-parser-parity.cjs wired into lint:ci; a lint script is not the test suite, so the exclusion stands. The test file keeps only the docs/CONTEXT-INDEX.json sync check. Blocker: the mutation test leaked its temp dir. Its callback took no `t`, so a failing assertion skipped the bare cleanup call. Now registered via t.after(), matching the convention adr-index-gate.test.cjs documents. Blocker: the example's own committed index carries the identical merge-race staleness this PR fixes for the production one, and nothing guarded it. Deliberately NOT fixed by wiring the example's --check into CI: that artifact bakes line numbers, so it re-drifts on any unrelated CONTEXT.md line shift -- exactly ADR-1671 open question 4 -- and would make CI routinely red. The new lint asserts the line-INDEPENDENT facts instead: count, class map, duplicate set, and every (id, value) pair. Proven non-vacuous both ways: mutating a value fails and names the id, mutating only a line number passes. Major: a real divergence the parity claim would have missed. Production rejects values containing an embedded CR, LF, U+2028 or U+2029; the example did not, so a value with an embedded lone CR was rejected by one copy and accepted by the other. Ported, and now covered by the parity table. Also, found while verifying rather than reported: malformed diagnostics covered only empty values. A doubled-dot id, a space in an id, and a lowercase-leading id were all dropped silently. That contradicts the module's own intent -- a typo should be diagnosable, and a space in an id is a likely one -- and predicates are contractually cited, so a silently vanished predicate is the failure mode that matters. Each rejection class now carries a named reason in both copies, while ordinary inline code still yields none. Trues up counts my own change staled: the example README and ADR-1671's prototype figures said 416 and 393/18 against a real 415/20/0. Closes #2944 * chore(#2944): backfill changeset PR number 2950 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
07603df8f2 |
fix(#2647): code-fixer worktree under .claude/worktrees/, not a hardcoded /tmp path (#2942)
* test(#2647): failing-first — fixer worktree path must be repo-relative not /tmp * fix(#2647): place code-fixer worktree under .claude/worktrees/, not /tmp The gsd-code-fixer agent hand-rolled its worktree at a hardcoded `/tmp/sv-${padded_phase}-reviewfix-XXXXXX` mktemp path. On Windows/Git Bash that landed OUTSIDE the project tree — outside the agent session's permission allowlist, so every Read inside the worktree prompted (~25/run) — and mktemp's MAX_PATH-avoidance substitute produced an un-removable `C:/mvwtNN` path. Place the worktree repo-relative under `.claude/worktrees/` (the same dir the harness-managed executor worktrees use: gitignored via `.claude/`, inside the session's permission scope), with a $$-PID + epoch suffix for concurrency uniqueness (replacing mktemp's XXXXXX). $main_repo is resolved the same way the cleanup tail already resolves it. Three sites updated: setup_worktree bash, concrete-steps prose, critical_rules. The #2990 `-b "$reviewfix_branch"` invariant is preserved (the folded test asserts it). Failing-first regression added to the #2990 suite in tests/agent-frontmatter.test.cjs. * test(#2647): update #2686 path assertion to expect .claude/worktrees/, not /tmp The #2686 regression test encoded the worktree location as a hardcoded `/tmp/sv-` path (matching sibling GSD agents at the time). #2647 showed that breaks Windows/Git Bash (worktree outside the project tree → permission prompts; mktemp MAX_PATH substitute un-removable). Update the #2686 path assertion to require the repo-relative `.claude/worktrees/` location and forbid `/tmp/sv-`. The #2686 isolation + cleanup assertions are unchanged. * fix(#2647): word-boundary wt= parse + ack the fixer growth vs next Two follow-ups to the #2647 GREEN run: - parseWtAssignments matched `prior_wt=` (no word boundary), polluting the set and tripping the repo-relative + concurrency-unique assertions. Anchor on (?:^|\s)wt= so only the real worktree-path assignment is captured. - emitted-attribution: gsd-code-fixer.md grew 1875 bytes vs origin/next. Update the emitted-drift-ack entry to attribute the #2647 worktree-path change (supersedes the prior #2825 attribution, whose growth is already in next). * fix(#2647): address review — validate padded_phase at the sink + tighten test Code-review + security-review both APPROVED with one actionable minor: padded_phase is interpolated into a worktree PATH and a git BRANCH NAME, but was only validated by the orchestrator (code-review-fix.md), not at the agent sink. The agent prompt is a literal bash contract any caller can spawn, so add a `[[ =~ ^[0-9]+(\.[0-9]+)?$ ]]` self-defense check rejecting traversal/shell metachars (defense-in-depth; not a present vuln — the only caller validates). Also tighten the concurrency-uniqueness test to require BOTH $$ AND $(date +%s) (either-alone was too lax per review). Update the emitted-drift-ack reason to cover the added validation growth. * changeset(#2647): code-fixer worktree under .claude/worktrees not /tmp * changeset(#2647): backfill PR number 2942 * chore(#2938): regenerate stale docs/CONTEXT-INDEX.json on next #2938 (#2928) updated the CONTEXT.md RULESET prose for the new per-PR emitted-drift-ack fragment mechanism (#2914) but shipped a CONTEXT-INDEX.json generated from the OLD prose. lint:generated-sync fails on every PR that rebases onto next after #2938 (the regen produces a 3-line diff bringing three RULESET entries — AGENT_SIZE_BUDGET, EMITTED_ATTRIBUTION, WORKFLOW_SIZE_BUDGET — in sync with the prose already on next). Mechanical regen via `node scripts/gen-context-index.cjs --write`; idempotent; surfaced by the #2647 rebase. No behavioral change. --------- Co-authored-by: sim <sim@users.noreply.github.com> |
||
|
Tom Boucher
|
05b170e448 |
chore(#2928): productionize the CONTEXT.md predicate fact-store and gate it in CI (#2938)
* feat(#2928): port CONTEXT.md predicate fact-store into the src seam Productionizes the ADR-1671 Option-E reference example as a real module: src/context-predicates.cts (parser + selector + index builder) compiled to gsd-core/bin/lib/, plus scripts/gen-context-index.cjs following the repo's --check/--write drift-guard idiom and wired into lint:generated-sync. Parser behavior is deliberately prototype-equivalent in this commit so the next commit's regression matrix binds to the real defects rather than to a missing module. Two locked design deviations from the prototype: - duplicates carry a count, not line numbers - the committed index carries no line field at all, resolving ADR-1671 open question 4: an artifact without line numbers cannot drift on a line shift, so promoting --check to a CI gate does not make it routinely red Also reconciles the one remaining duplicate predicate ID (RULESET.WORKFLOW_MARKDOWN.FENCES was declared twice; the non-MD040 wording is removed) so the gate can land fail-closed on duplicates. Refs #1671 * test(#2928): failing-first matrix for the predicate fact-store Adds the regression matrix from the phase test plan: parser declaration forms, fence and comment regions, ID/value grammar boundaries at limit-1/limit/limit+1, CRLF fidelity, duplicate detection, the drift-guard CLI, the selector query surface, and four document-shaped fast-check properties. Seven rows are RED for behavioral reasons against the ported parser: indented-bare, star-list, plus-list and numbered-list declaration forms are dropped; a tilde fence and a four-backtick fence containing a shorter fence are not skipped; and a multi-line HTML comment is parsed as live. Eleven selector rows are RED because the query surface is not wired yet. Negative fixtures come from real repo documents that predate the grammar (CONTEXT.md, CONTRIBUTING.md's fenced env-assignment examples) per the fixture-provenance rule, and the property generators are document-shaped rather than seeded from our own serializer. Refs #1671 * fix(#2928): consume the shared fence scanner, relocate the index, wire the selector Drives the failing-first matrix green. Parser: replaces the ported naive triple-backtick toggle with the shared markdown-sectionizer fence engine. scanFencedBlocks and FencedBlockRecord gain an export keyword — the only change to that module, which has 71 upstream dependents — because it already returns line-indexed spans, which is exactly what a line-reporting parser needs. It also already documents itself as the second copy of the fence state machine pending consolidation; adding a third copy here would have been the generative-fix divergence this repo warns about. A parity suite now pins predicate fence-skipping against that scanner across eight fence shapes. HTML-comment skipping stays local because the sectionizer has no comment scanner. Declaration forms widen to indented-bare, star, plus and numbered list items. Index location: docs/CONTEXT-INDEX.json, not a module under bin/lib. The remote matrix run caught the original choice — a committed .cjs there ships ~120KB of CONTEXT.md prose into a runtime module, and two content guards fired truthfully on it (a leaked .claude install path, and four hardcoded package-name literals). Neither guard was allowlisted; the artifact moved instead, mirroring docs/INVENTORY-MANIFEST.json. Nothing at runtime needs to require it — it is a drift-detection artifact, so the selector parses CONTEXT.md live and is always current. Generator: adds a frozen REASON enum and --check --json so the gate's outcome is asserted structurally instead of by matching prose, and --context-path/--index-path so tests drive the real CLI against a temp tree with no filesystem monkeypatching. Selector: gsd_run query context-predicates with --class/--prefix/--contains, structured output carrying a matched count, own-property guards, and no project-root resolution. Registering it exposed that the query dispatch table and the usage string had drifted: a new parity test found 20 routed commands missing from the usage list, all added here rather than deferred. Refs #1671 * test(#2928): lock the newly-public scanFencedBlocks contract Exporting scanFencedBlocks made it public API for the first time, so it needs its own contract test independent of the consumer that motivated the export. Memtrace's co-change analysis flagged the gap: this suite changes together with markdown-sectionizer.cts 8 times in 90 days and was absent from the diff. Covers the documented rules: 0-based indices, -1 for an unterminated fence, the same-char/>=length/no-trailing-text closer rule, a shorter fence inside a longer one staying content, CommonMark 4.5 backtick-in-info-string, and <=3-space indent tolerance. Refs #1671 * fix(#2928): address both isolated review passes Two independent reviewers (correctness axis and security axis, neither the author) found seven findings. All are fixed here with regression tests; none deferred. BLOCKER — comment-blind fence scanning caused silent, permanent predicate loss. The HTML-comment scan and the fence scan ran as two independent passes, and the fence scanner is comment-blind, so a fence delimiter inside an HTML comment with no later close read as an unterminated fence and skipped every remaining line to EOF. Worse, the drift-guard could not catch it: it diffs against a baseline produced by the same corrupted parse. The two constructs now interleave in a single pass so each suppresses the other's boundary detection while active, covered in both directions. The parity suite still binds this scanner to markdown-sectionizer's for comment-free documents, so the two cannot diverge unnoticed. BLOCKER — the selector was not consumed anywhere, leaving the phase's acceptance criterion unmet. Now wired into the pre-work predicate-citation step in contributor-standards, which is the repo's actual brief-assembly path; no code-level brief assembler exists to wire into. MAJOR — ReDoS with an unauthenticated CI-hang exploit. The predicate-id regex nested a dot-containing character class inside a dot-prefixed repeat, so N consecutive dots had exponentially many partitions: 40 dots took 565ms and growth was exponential. CI runs this parser over a pull request's own CONTEXT.md, so any contributor could have hung a shared runner with one line. Replaced with linear per-segment validation. Doubled-dot ids are now rejected; the real document contains none. MAJOR — the duplicate-id gate had only ever been proven on synthetic fixtures. A test now re-inserts the exact line this branch removed and asserts the real generator names it. MAJOR — --check together with --write silently let write win, turning the gate into a writer; a missing path value resolved to the cwd and leaked an EISDIR stack trace. Both are now clean usage errors. MINOR — the hoisted skip-list was exported as a live mutable Set; replaced with a read-only predicate. MINOR — flag-shaped selector values were unmatchable; the inline --flag=value form now provides the escape hatch. Refs #1671 * chore(#2928): backfill changeset PR number 2938 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
c043f2946c |
fix(#2914): per-PR ack fragments instead of one shared mutable file (#2923)
* fix(#2914): never persist a spent emitted-drift ack on next tests/emitted-drift-ack.json held 34 spent #2834 entries merged via #2900. Every entry is scoped to the diff that introduced it (#2789), so once merged to next it is at the base by definition -- spent and inert. Its presence is still load-bearing though: each PR rewrites the paths map wholesale, making a persistent base copy a shared cell. Five of six conflicting PRs in the open queue collided on this file and nothing else. Deletes the stale document and adds a push-to-next guard asserting it stays absent. The guard is deliberately NOT wired into lint:ci -- a PR-lane check against the base is the #2768 shape #2789 exists to end. Closes #2914 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2914): backfill changeset pr number Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2914): per-PR ack fragments instead of one shared mutable file The emitted-drift acknowledgment lived in a single tests/emitted-drift-ack.json whose paths map every PR rewrote wholesale. That is a shared mutable cell: any two PRs needing an ack edit the same lines and conflict. Five of six conflicting PRs in the open queue collided on this file and nothing else. Acks now live as per-PR fragments under tests/emitted-drift-acks/, the same shape .changeset/ already uses to solve this exact problem. Two PRs pick different filenames, so they cannot collide, and fragments lingering on next are harmless rather than toxic. The legacy file's 35 entries are MIGRATED into a fragment, not deleted. An earlier delete-only attempt failed verification twice: the ratchet lost the spec-phase.md acknowledgment from #2779 and reported a 10-byte growth with no ack. Relocating preserves every acknowledgment. The legacy single file is still READ (unioned with the fragments) because five open PRs carry it; dropping support would break all of them. A duplicate path key across sources is a hard error, never last-wins. The push-to-next guard is retargeted accordingly: it now asserts only that the legacy SHARED file never reappears on next. Fragments may persist harmlessly. Closes #2914 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
d2d2f7c088 |
fix(#2848): non-Latin titles no longer produce empty slugs (Cyrillic transliteration) (#2934)
* test(#2848): add failing-first regression for non-Latin slug transliteration generateSlugInternal and slugify both strip non-ASCII chars with no transliteration step, so an all-Cyrillic title reduces to an empty slug. 12-row matrix: Cyrillic regression (both impls), Latin negative control, multi-letter mappings, soft/hard sign drops, Ukrainian extras, null contract, CJK unaffected, mixed scripts, truncation parity, slugify's distinct no-truncate contract. * fix(#2848): transliterate Cyrillic titles to ASCII before slug strip Both generateSlugInternal (src/core-utils.cts) and slugify (src/gsd2-import.cts) stripped non-ASCII with no transliteration, so an all-Cyrillic title reduced to an empty slug, producing unnamed phase directories (01-) and empty milestone_slug init JSON. Add a shared transliterateForSlug primitive (core-utils) covering Russian + the reported Ukrainian/Belarusian extras (і ї є ґ ў), with multi-letter mappings (ж→zh ч→ch ш→sh щ→sch ю→yu я→ya) and dropped soft/hard signs (ъ ь). It runs BEFORE the existing ASCII filter, so Latin-script text hits zero map entries and is byte-for-byte unchanged (negative control). slugify consumes the shared primitive, preserving its distinct single hyphen-strip + no-truncation contract. CJK/unmapped scripts keep the existing strip-to-ASCII behavior. Also corrects two test assertions to match the chosen й→y mapping and the б→b (not bie) transliteration. * changeset(#2848): Fixed — non-Latin slug transliteration * changeset(#2848): backfill PR number 2934 --------- Co-authored-by: sim <sim@local> |
||
|
Rezolv
|
76b7d73039 |
fix(#2733): route gate-passed spec-phase paths into the probe steps (#2779)
* fix(#2733): route gate-passed spec-phase paths into the probe steps All four gate-passed transitions in spec-phase.md said "Jump to Step 6", textually bypassing the mandatory Step 5.5 edge-completeness and Step 5.6 prohibition-completeness probes. Steps 5.5/5.6 were spliced between Step 5 and Step 6 by two later feature commits and the pre-existing jumps were never re-pointed, so no jump instruction in the file reached Step 5.5 at all and both probes were unreachable dead prose. Re-point the four gate-passed jumps (lines 129, 162, 168, 170) to Step 5.5. Control then flows 5.5 -> 5.6 -> 6 as the probes' own preconditions prescribe. The max-rounds "write anyway" bypasses and the probes' own "proceed to Step 6" exits are deliberately unchanged. Add tests/spec-phase-probe-reachability.test.cjs, which derives the mandatory probe steps from the file's own headings rather than hardcoding 5.5/5.6, so a future spliced-in probe step is covered without editing the test. It also locks the two coupled constraints: the max-rounds bypass must not be redirected into a probe, and each probe must keep its own onward exit. The existing probe contract tests are untouched and still pass; both scope from the "## Step 5.5"/"## Step 5.6" heading onward and were structurally incapable of observing the upstream jump text. * chore(changeset): Fixed fragment for #2779 (spec-phase probe reachability) * fix(#2733): route Step 5.5's own soft gate into Step 5.6 Round-1 review blocker. The four upstream gate-passed jumps were re-pointed to Step 5.5, but Step 5.5's own terminal soft gate at :305 still read "proceed to Step 6" - so the COMMON path (all applicable edges resolved) skipped the prohibition-completeness probe outright. Same defect class as the four this PR already fixed, on the success path of the very step being fixed: the SPEC shipped with an empty Prohibitions section instead of an empty Edge Coverage one. Its sibling at :393 is byte-identical yet correct, because Step 6 genuinely follows Step 5.6. Position, not phrasing, is the discriminator. The guard could not see it: the transition matcher keyed only on the literal "Jump to Step", and :305 says "proceed to Step". Widened it to a verb alternation (jump/proceed/continue/go/return/skip + "to Step N", case-insensitive) and renamed it TRANSITION_RE to match what it now models. This makes the file's own docstring promise - that a future spliced-in probe is covered without editing the test - true for a step whose exit is worded differently. Verified no false positives: the two pre-existing "continue to Step 3/4" transitions are upstream of both probes but target pre-probe steps, and the max-rounds bypass block contains no step transitions at all. Fail-first verified before fixing :305 - with the widened matcher against the unfixed workflow the guard fails naming exactly "spec-phase.md:305 jumps to Step 6, skipping mandatory Step 5.6", 4 pass / 1 fail; after the fix, 5/5. The two sibling probe contract tests stay 16/16. Also from review: - STEP_HEADING_RE gains an explicit \r? before $. Without it, on a CRLF checkout `.` stops before the \r and the unanchored $ fails to match, yielding ZERO steps and vacuously passing every assertion in the file. Not live today (.gitattributes forces eol=lf) but this repo has a recurring CRLF-regex bug class, so the guard no longer leans on it. - allow-test-rule category corrected to source-text-is-the-product; the previous runtime-contract-is-the-product is not one of the six recognized categories (CONTRIBUTING.md:609-619). - changeset body given the documented bold-lead-in form. - emitted-drift ack reason updated: +8 -> +10 bytes across five transitions (31987 -> 31997), DEFAULT tier, cap 40960. --------- Co-authored-by: CI Rebase Check <ci@gsd-redux> Co-authored-by: Tom Boucher <trekkie@nomorestars.com> |
||
|
Tom Boucher
|
d49a7d0c4d |
fix(#2853): roadmap.update-plan-progress preserves hand-written annotations (#2916)
* test(#2853): add failing-first regression for plan-progress annotation preservation The count-bump regex's trailing [^\n]+ swallowed the whole Plans line and the replacement wrote back only the regenerated count, deleting any hand-written annotation after it. 8-row matrix covers bold/plain forms, bare template form, executed path, CRLF, and idempotency. * fix(#2853): preserve hand-written annotations in roadmap plan-progress bump The count-bump regex's trailing [^\n]+ swallowed the entire Plans line and the replacement wrote back only the regenerated count, deleting any hand-written prose after the count (e.g. a gap-closure annotation). The verb owns the count token only. Capture the existing count token ($2) and the trailing line text ($3), and rebuild the line as <label><new count><surviving text>. Trailing text is preserved ONLY when a real count token preceded it, so the fresh-template bracketed placeholder (`[Number of plans…]`) is still replaced cleanly rather than glued after the count (pre-#2853 behaviour on the template path preserved). CRLF \r is preserved via [^\r\n]. Widens replaceInCurrentMilestone to accept a replacement callback (needed to branch on whether the count group matched). The bare Plans: checklist header is still skipped — the lazy match lands on the summary line first and a count-less bare header yields no count to anchor preservation to. * changeset(#2853): backfill PR number 2916 --------- Co-authored-by: Test <test@example.com> Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
90771ddf02 |
enh(#2904): add a reviewer entry type so third-party reviewer lanes are discoverable (#2912)
* feat(#2904): add a `reviewer` entry type so third-party reviewer lanes are discoverable ADR-2782 made a reviewer lane installable by a third party, but neither discoverability catalog could hold one. The Community Capability Registry requires a non-empty `loopExtensionPoints` and forbids a lane from declaring any hook kind, so a `role: "reviewer"` entry is unsatisfiable by construction; the EoS Registry is for ADR-1239 host integrations, which a lane is not. Adds a third catalog — `docs/registries/reviewers.json` → `docs/registries/reviewer-registry.md` — whose `interactions` describes the lane: slug, flags, transport, evidenceClass, reviewsSection, requiresBinaries, configKeys, runtimeCompat. The lane vocabulary is a hand-written mirror of `capability-validator.cjs` (the same pattern as `AXES` mirroring `HOST_INTEGRATION_AXES`), with parity enforced by tests/registry-reviewer-parity.test.cjs. `slug` deliberately uses the runtime `LANE_SLUG_RE` grammar rather than the registry's kebab-only `id` rule, so real lanes (`lm_studio`, `4o-mini`) are not rejected. Two binary type branches became three-way Map dispatch. Both now fail loudly on an unrecognized type instead of silently treating it as a capability — `renderMarkdown` in particular writes a committed catalog file, so a silent wrong-title render was the worst failure mode available. Also fixed while here: `gen-registry.cjs` parsed source JSON with no error handling, so a malformed or non-array `capabilities.json` surfaced as a raw SyntaxError/TypeError instead of an actionable CLI error. Closes #2904 * fix(#2904): bound and sanitize untrusted registry `interactions` strings Review findings from the pre-PR passes. Security (isolated pass): `interactions` string fields reached the generated, committed Markdown catalog with no control-character check and no length bound. A `reviewsSection` carrying ESC and a `requiresBinaries` element carrying NUL plus 5000 characters validated clean and landed verbatim in the rendered page — `mdInline` escapes Markdown metacharacters and collapses CRLF, but nothing else. The identical gap already existed on the capability type's `configKeys`/`requires`/`runtimeCompat`/`produces`/`consumes`, so it is fixed there too rather than inherited into a third type. `hasDisallowedControlChar` is lifted to module scope so exactly one implementation exists, and a shared `validateStringArrayField` enforces control-character rejection, a 200-character element cap and a 50-element array cap for both types. Correctness (standards pass): `renderMarkdown`'s per-entry summary builder was still an if/else-if chain whose final `else` was the capability branch — the one per-type dispatch point this change had not converted, and the same silent fallthrough it removes elsewhere. It now lives in `RENDER_META` alongside the title, so a fourth type cannot silently inherit capability's rendering. All three types' rendered output is byte-identical to before the refactor. Also corrects a test comment that still claimed the reviewer suites were failing-first against an unmodified module. * chore(#2904): backfill changeset PR number (#2912) |
||
|
Tom Boucher
|
42f4f184c0 |
fix(#2844): verify-summary ignores future/prose path mentions; resolves project root (#2910)
* fix(#2844): verify-summary binds file-claim extraction to a creation-claim context verify-summary's Pattern 1 matched any backticked path-like token with no context check, so a prose mention of a future deliverable (`shared/types.ts` in a 'next phase will add…' sentence) was checked for existence and its absence failed the verdict on a healthy phase. #2685 added shape filtering but no context check. - src/verify.cts: both extraction patterns now require a claim label on the line (Created/Modified/Added/Updated/Edited/key-files). A bare prose mention no longer matches; genuine labeled claims still do. - gsd-core/bin/gsd-tools.cjs: remove 'verify-summary' from SKIP_ROOT_RESOLUTION so relative claim paths resolve against the project root, not the raw cwd (subdirectory invocation no longer manufactures missing files). Regression tests: prose mention not treated as a claim; prose-only SUMMARY passes; absent claimed file still fails. * chore(#2844): backfill changeset PR 2910 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
39dbe5e0f5 |
fix(#2843): findProjectRoot stops at git-repo boundaries, not just MAX_DEPTH (#2909)
* fix(#2843): findProjectRoot stops at git-repo boundaries, not just MAX_DEPTH findProjectRoot's heuristic (3) used isInsideGitRepo(parent), which only checked 'does SOME .git exist between start and the ancestor' — it never verified the .git was co-located with / bounded the trusted .planning/. A nested child repo (own .git, no .planning) under an ancestor GSD project satisfied the check, so resolution silently crossed into the ancestor project (wrong identity, exit 0). Add nearestGitRoot(from, upTo) (fs-walk, no spawn) and use it in heuristics (3) and (4): if the caller is inside its own nested repo whose root is strictly below the candidate ancestor, do not return that ancestor. The plain-descendant (#1414), co-located .git+.planning, and sub_repos/multiRepo cases are unchanged. Regression test: a nested child .git under an ancestor .planning no longer resolves to the ancestor; the co-located single-repo case still resolves. * chore(#2843): backfill changeset PR 2909 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
5d0fd4dc53 |
fix(#2825): gsd-code-fixer honors workflow.use_worktrees; never rm -rf a reparse point (#2905)
* fix(#2825): gsd-code-fixer honors workflow.use_worktrees; never rm -rf a reparse point gsd-code-fixer was the only writer that hand-rolled a git worktree inside the agent prompt and the only one that never read workflow.use_worktrees. With the setting explicitly false, --fix still created worktrees; the fresh worktree had no node_modules, so runs improvised a teardown whose `rm -rf` followed a Windows junction into the REAL node_modules (silent data loss, 3x observed). Defect 1: gate setup_worktree + its cleanup tail on workflow.use_worktrees (same gsd_run query config-get read the four sibling workflows use). When false: edit/commit in the main checkout (wt='.', no temp branch, no sentinel, no cleanup). Defect 2: forbid rm -rf on a possible reparse point in the spec — never fall through to a destructive remove; on failure, stop and surface the error. Defect 3: REVIEW-FIX records where verification ran (main checkout vs worktree). The transactional worktree path (#2839/#2990/#2686) is unchanged when worktrees are enabled. Docs-parity guards in tests/code-review.test.cjs bind the spec to the fix. * chore(#2825): backfill changeset PR 2905 * fix(#2825): drop stale agents/gsd-code-fixer.toml ack + spent entries (emitted-attribution) gsd-test failed: agents/gsd-code-fixer.toml is a STALE ack here — that TOML was changed by #2834 (now in next), not this PR. The 4 base-carried acks (autonomous/discuss-phase-assumptions/next/plan-phase) are spent/inert. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
00c859fae5 |
fix(#2834): write defaults.json before agent TOML generation on clean Codex install (#2900)
* fix(#2834): write defaults.json before agent TOML generation on clean Codex install Extracted writeNonClaudeDefaults(runtime) and called it BEFORE installCodexConfig so the runtime-aware model resolver has resolve_model_ids=omit + runtime=codex in ~/.gsd/defaults.json before agent TOMLs are generated. Pre-fix, a clean first Codex install generated TOMLs with no model fields (the resolver didn't know the runtime); a second run fixed it. The original inline defaults-write block (which ran AFTER agent generation) is replaced by the earlier function call (idempotent). * chore(#2834): changeset fragment * fix+test(#2834): acknowledge codex TOML drift (emitted-attribution) + fix test comment window The emitted-attribution gate flags 19 codex agent TOMLs that now carry model-routing fields (the fix's correct effect) but can't link them to a .md or src/ change (the fix is in bin/install.js ordering). Acknowledge the drift in emitted-drift-ack.json. Fix the test's comment-detection window (300 chars to capture the #2834 rationale). * fix(#2834): ack remaining 15 codex TOML drift paths * chore(#2834): backfill changeset PR number (2900) * chore(#2834): ack code-review.md growth from concurrent merge (rebase pickup) * fix(#2834): remove stale code-review.md ack (emitted-attribution failure) CI failed: 'differential attribution over the real tree' — the code-review.md ack added in 6e0b4b3b8 ('ack code-review.md growth from concurrent merge') is STALE: this PR's diff does not touch code-review.md (only bin/install.js + tests + changeset), so the ack explains growth that isn't here. The base already absorbed the concurrent code-review.md growth; the ack is inert here and the gate flags it as stale. Remove it. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
9f567a1627 |
fix(#2788): recover Gaps Found rows; mark-complete no longer false-succeeds on a rejected row (#2902)
* fix(#2788): recover Gaps Found rows; mark-complete no longer false-succeeds on a rejected row Two coupled defects in the requirement traceability state machine: Defect 1 (terminal state): requirements revert-phase (the gaps_found response) left a row at 'Gaps Found' with no inverse — neither mark-complete's /^pending$/i guard nor the phase-complete reconcile's /^(?:pending|in progress)$/i accepted it, so a single failed verification stranded every requirement permanently and blocked the milestone. Widen both guards to accept 'gaps found' so a genuinely-satisfied stranded row reaches Complete again. Defect 2 (false success): mark-complete ORed checkboxHit || tableHit for 'updated', so on a Gaps Found row it flipped the checkbox but could not move the row, yet reported updated:true. When a traceability table has a row for an ID, gate 'updated' on the row moving (tableHit) — a checkbox-only flip on a table-bearing file no longer lies. The #2140 table_unmatched path (no row for the ID) is preserved. * chore(#2788): backfill changeset PR 2902 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
79ed181ec0 |
fix(#2667): run-with-timeout mediates .cmd/.bat spawns on Windows (CVE-2024-27980); fallow pre-pass names failure kind (#2897)
* fix(#2667): mediate .cmd/.bat/.exe spawns on Windows; split fallow pre-pass failure diagnostic run-with-timeout spawned .cmd/.bat/.exe commands without shell:true on Windows, tripping Node's CVE-2024-27980 EINVAL (April 2024 security hardening). The fallow structural pre-pass then no-op'd silently — a hard execution failure read the same as 'optional dependency absent'. (A) gsd-core/bin/gsd-tools.cjs runWithTimeout: gate shell:true on (win32 && command ends in .cmd/.bat/.exe). Narrow by design — never fires for the 7 `bash -c` callers (command is `bash`, no such suffix), so the recorded no-shell-for-argv-array security contract (DEFECT.UNBOUNDED-SUBPROCESS) is preserved; cmdArgs stays an array. POSIX untouched. (B) code-review.md fallow pre-pass: name the failure KIND (timeout / spawn failure / crash / not-found) so a Windows .cmd spawn failure is not mistaken for an absent binary. Regression test in tests/run-with-timeout.test.cjs gated to win32 (.cmd/.bat/.exe shims run with exit 0 + non-empty stdout; pre-fix EINVAL → exit 125/empty). POSIX negative-space test guards the unchanged bash -c callers. * chore(#2667): changeset fragment * chore(#2667): backfill changeset PR 2897 + correct body (cmd.exe array, not shell:true) * fix(#2667): exclude .exe from the win32 spawn-mediation gate; ack code-review.md growth CI caught two failures on the first push: 1. windows-24: 'exits 124 when the wall-clock budget is exceeded' regressed. The gate matched .exe, so the HANG command (node.exe -e 'setTimeout(...)') was wrapped in 'cmd.exe /c node.exe ...' — the wrapped child escaped the timeout cap's process-group reap (exit 124 never fired; hit the 30s harness backstop) AND cmd.exe risked mis-parsing the -e script arg. .exe is INTENTIONALLY excluded now: real PE executables spawn fine directly; only .cmd/.bat are the CVE-2024-27980 EINVAL cases. The .exe test becomes a negative-space test (node.exe spawned directly, exit 0). 2. ubuntu-22: emitted-attribution — code-review.md grew 1177 bytes from the #2667 fallow pre-pass failure-KIND case statement; acknowledge it. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
6e0bc50142 |
fix(#2666): code-review scopes root-level + extensionless build files, cross-checks against git diff (#2895)
* test(#2666): add regression + docs-parity guards for code-review file scoper The Tier-2 SUMMARY.md extractor dropped every repository-root file (no `/`) and every extensionless build file (Dockerfile/Makefile/etc.) via an AND-joined predicate. Adds behavioral tests against the pure-function mirror plus docs-parity structural guards that bind the shipped workflow .md to the fix. RED: the docs-parity guards fail against the pre-fix shipped predicate. * fix(#2666): accept root-level + extensionless build files in code-review scope; intersect-and-warn Two coordinated edits to gsd-core/workflows/code-review.md compute_file_scope: (A) Tier-2 SUMMARY extractor: replace the AND-joined predicate `/\\//.test(raw) && /\\.[A-Za-z0-9]+$/.test(raw)` (which required BOTH a directory separator AND a trailing extension, silently dropping every root-level file and every extensionless build file) with a relaxed predicate that accepts any path with a trailing extension OR a known extensionless build basename (Dockerfile/Containerfile/Makefile/Justfile/Procfile). (B) Tier-3: convert the eq-zero git-diff gate into an intersect-and-warn — whenever a reliable diff base is available, cross-check the SUMMARY scope against `git diff --name-only` and warn about (then add) any changed files the SUMMARY extractor did not surface. Portable (bash 3.2, no associative arrays) so a partial SUMMARY result can no longer silently ship an incomplete review scope. * chore(#2666): changeset fragment * fix(#2666): use exact whole-line matching (grep -Fxq) in Tier-3 cross-check Adversarial review found the unanchored `case "$IN_SCOPE" in *"$file"$\\n*` substring membership test would false-match: a root-level `Dockerfile` in the diff substring-matches an already-scoped `docker/Dockerfile`, silently skipping it — reintroducing the exact class of silent-scope-loss bug this PR fixes. Switch to `grep -Fxq` (exact whole-line match). Add docs-parity guard for exact matching + the basename-collision regression. * fix(#2666): resolve gsd-test failures — paraphrase predicate in comment, ack code-review.md growth gsd-test caught 3 issues on 7469c3f22: 1. The docs-parity guard fired on the .md COMMENT which restated the buggy predicate verbatim — paraphrase the comment so it no longer contains the exact string the guard detects. 2. Cascade subtest failure from #1. 3. emitted-attribution: code-review.md grew 2814 bytes — acknowledge the deliberate #2666 growth in tests/emitted-drift-ack.json. * chore(#2666): backfill changeset PR number 2895 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
f093738412 |
fix(#2891): normalize emitted version against the measured tree, not the measuring repo (#2894)
* fix(#2891): normalize emitted version against the measured tree, not the measuring repo buildParityManifest normalized the install-time {{GSD_VERSION}} stamp using PKG_VERSION, bound at module load from the MEASURING repo's package.json. Since #2767, currentManifests({repoRoot}) measures a DIFFERENT checkout, so during a release cut the baseline worktree (origin/next, 1.8.0) was normalized with the current tree's version (1.9.0) and its literal 1.8.0 stamp survived into the hash. All 364 emitted hook paths diverged and the differential attribution gate hard-failed every finalize/rc run. Normalize against the version of the tree that PRODUCED the emitted output: buildParityManifest takes an explicit pkgVersion, and currentManifests resolves it from the measured tree via a new fail-closed measuredPackageVersion(). * chore(#2891): backfill changeset pr number (#2894) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
cd5b8643ae |
fix(#2828): state sync reports correct total_phases on a flat unmilestoned roadmap (#2892)
* fix+test(#2828): total_phases uses roadmap count on flat unmilestoned roadmap The read-path disk-scan cache fell back to phaseDirs.length (1) when milestoneBounded was false, even though roadmapPhaseCount (6) was correct for a flat roadmap (no sibling milestones to conflate). Use roadmapPhaseCount as the floor when > 0, matching the write-path (cmdStateSync) which already did this. The milestoneBounded flag still flows to milestoneUnbounded for the percent-skip (#1761 guard preserved). Regression test asserts state-sync writes progress.total_phases:6 for a flat 6-phase roadmap + 1 phase dir. * chore(#2828): changeset fragment * test(#2828): add negative-space coverage (Math.max floor mutant + no-roadmap fallback) — review findings The 6-phase test alone couldn't kill a Math.max-dropping mutant (1<6). Add: a 3-phase-dir/2-roadmap-phase case proving Math.max(dirs,count) floor; a no-roadmap case proving phaseDirs.length fallback. * fix(#2828): refine — distinguish flat unmilestoned from milestoned-unbounded (preserve #1761) The first-pass fix (roadmapPhaseCount > 0 always) re-broke #1761: a milestoned- unbounded roadmap (asserted milestone not among existing version headings) conflated sibling milestones (8 = 4+4). Refine with a hasMilestoneSectioning discriminator: ^#{2,3}(?!Phase) detects non-Phase h2/h3 milestone section headings. A FLAT roadmap (only ### Phase headings + a # title) has none → safe to use roadmapPhaseCount; a SECTIONED-but-unbounded roadmap has them → fall back to phaseDirs.length (#1761). Verified both cases locally (flat→6, sectioned-unbounded→1). * test(#2828): remove two fragile negative-space tests (phase-dir scanner internals) The Math.max-floor and no-roadmap tests made assumptions about the phase-dir scanner's internals (which dirs count as 'realized') that didn't hold. The core regression test (6-phase flat → total_phases:6) plus the existing #1761 conflation tests (which the refined fix preserves) provide sufficient coverage. * chore(#2828): backfill changeset PR number 2892 * fix(#2828): replace ReDoS-prone regex in regression test with line-by-line parse CodeQL flagged the nested-quantifier regex (`(?:[ \t]+\w+:.+\r?\n?)*?`) in tests/issue-2828-flat-roadmap-total-phases.test.cjs as a high-severity catastrophic-backtracking risk. Rewrite the STATE.md progress.total_phases extraction as a ReDoS-safe line-by-line block walk. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
54cb4145bf |
fix(#2765): bump brace-expansion to patched 1.1.18/5.0.9 (high-severity devDep advisory) (#2888)
* fix(#2765): bump brace-expansion to patched 1.1.18/5.0.9 (high-severity devDep advisory) npm audit fix (non-breaking) bumps the lockfile: brace-expansion 1.1.15→1.1.18 (eslint-nested via minimatch@3.x) and 5.0.6→5.0.9 (stryker-nested). Both 1.1.18 and 5.0.9 were published 2026-07-30 as the patch backports for GHSA-3jxr-9vmj-r5cp / GHSA-mh99-v99m-4gvg (range <=5.0.7). No overrides needed (in-range bump), no major bumps, no --force. Production (npm audit --omit=dev) unaffected (devDep only). Add a structural test pinning the installed versions so the bump can't silently regress. * chore(#2765): changeset fragment * fix(#2765): correct changeset issue ref + parse patch version as number (review findings) - changeset cited #2762 (typo) — fix to (#2765). - test compared v.split('.')[2] as a string (false-pass for 1.1.9) — parse all segments as Number. * chore(#2765): backfill changeset PR number (2888) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
4bd6fb066b |
chore(#2880): close ADR-2143 deployment misses — table-regex fingerprint + state-document seam migration (#2889)
* chore(#2880): close ADR-2143 seam misses — widen table-regex fingerprint, migrate state-document onto the seam The no-adhoc-markdown-parsing rule matched only a negated class whose sole member was a pipe ([^|]), so the stricter and more common [^|\n] spelling evaded it entirely -- src/state-document.cts hand-rolled exactly that shape and linted clean. Widen the fingerprint to any negated class excluding a pipe, which is the ADR-2143 section 7 prohibition as written. With the rule fixed, state-document.cts goes red. Replace tableRowPattern with locateFieldRow: a line scan using the markdown-table seam's splitTableRow for cell semantics, returning the value cell's byte range, and splice that range instead of running a whole-document content.replace. An edit now physically cannot cross a row boundary (section 4). Behavior is frozen -- stateReplaceField has 79 dependents across 5 command processes. Characterization tests lock all 14 table-branch rows plus CRLF, extract round-trip and the withFallback caller shape; a fast-check property asserts every non-target line stays byte-identical. Refs #2880, epic #2143 * fix(#2880): address adversarial review — lone-CR rows, field-name padding, quadratic scan, over-broad fingerprint Isolated adversarial review found four defects in the first commit. 1. locateFieldRow split lines on \n only. JS treats a lone \r as a line terminator, so the regex it replaced matched rows separated by bare CR. "| Phase | 3 |\r| Other | 9 |" returned 3 before and null after. Now CR, LF and CRLF are all terminators, byte offsets unchanged. 2. The field name was normalised with trim().toLowerCase(). The old regex embedded it verbatim, so its whitespace had to be absorbed by the row's own padding -- and because the group is a literal-character match rather than a whitespace class, a tab-padded cell does not accept a space-padded name. Replaced with an offset-aligned search reproducing the original backtracking exactly. 3. The widened fingerprint regex had two unbounded [^\]]* around an optional and ran quadratically over every regex source in every linted file: 256000 chars took 23 seconds. Replaced with a single-pass scanner that never rescans; the same input is now ~1ms. 4. The fingerprint also matched non-table idioms such as [^\s|] and [^"|]. Narrowed to a class excluding the pipe plus only \n, \r or \t. Differential fuzz against origin/next: 20000 cases, 0 mismatches. Refs #2880, epic #2143 * test(#2880): drop wall-clock assertion from the ReDoS regression guard local/no-elapsed-assertion flagged the elapsed-time check, and CLAUDE.md bans timing assertions outright as flaky. The 256000-char input stays as the regression guard for the quadratic scan; correctness of the verdict is what is asserted. If the quadratic path returns, the test stops completing and surfaces as a suite timeout rather than a silent pass. Also adds the changeset fragment for #2880. Refs #2880 * fix(#2880): spec-correct case folding, property tests, naming Code-review findings. The field-name comparison used toLowerCase(). The regex it replaced used /i WITHOUT /u, and ECMAScript Canonicalize deliberately does not fold a non-ASCII character onto an ASCII one -- KELVIN SIGN U+212A matched ASCII K where the old code returned null. Replaced with spec-correct Canonicalize, including the multi-character uppercase case (eszett -> SS), which a naive uppercase comparison also gets wrong. Added the fast-check property tests CLAUDE.md requires for parsers: one for the negated-class scanner, one for the field-name fold semantics, each against an independent reference implementation. Both reference impls failed on first run against real bugs, so neither property is vacuous. Renamed p2/p3 to name the exactly-three-pipes invariant, and reduced a duplicated comment to a cross-reference. Differential fuzz vs origin/next: 20000 runs, 0 mismatches, with the harness proven to discriminate the KELVIN case. Refs #2880 * chore(#2880): backfill changeset PR number (#2889) * docs(#2890): correct the local ESLint plugin path in CONTEXT.md CONTEXT.md named the local AST-rule plugin directory as scripts/eslint-rules/, which does not exist. The real location is eslint-rules/ at the repo root -- what eslint.config.mjs actually imports -- and CONTEXT.md's own later entry already says so explicitly, so the file disagreed with itself. Found by a line-by-line audit of all 1036 lines against the live graph; this was the only confirmed inaccuracy. Closes #2890 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
7372d99a26 |
enhance(#2800): derive reviewer flag lists and gate reviewer lane docs across locales (#2882)
* chore(#2800): derive reviewer flag lists and gate reviewer lane docs across locales The reviewer lane roster was hand-enumerated across five documentation surfaces and three workflow files that had drifted apart: --kimi-code was missing from all four translated COMMANDS.md mirrors, --coderabbit from every workflow forwarding list, and --antigravity from FEATURES.md. Adds checkReviewerDocsParity, a second pure gate deliberately separate from checkReviewerLaneParity so a stale doc cannot make the runtime checker look red. Workflows now derive their flag lists from a new review-lane flags query instead of hand-enumerating them, which also retires the unanchored grep that matched --agy inside --antigravity. Documents the previously absent reviewer body and hostBehaviors field in the capability manifest reference. Closes #2800 Closes #2781 Closes #2272 * fix(#2800): key the docs parity table arm on first-cell position Review found the flag arm was file-scoped, so the forwarding row that lists every flag in its third cell satisfied it on its own. Deleting a lane's own reviewer-table row -- the #2781 regression this gate exists to prevent -- therefore passed undetected. Arm 4 keys on the FIRST table cell, which separates a lane row from the forwarding row structurally and in every locale. Regression test included. * fix(#2800): shape-filter the flags subcommand output All three consumers read review-lane flags through an unquoted command substitution so the output word-splits into loop items. Phase 2 admits third-party overlay lanes, so an overlay flag containing whitespace would inject a second loop item and one containing a glob would expand against the cwd. Emit only well-formed flags so neither reaches the shell. * fix(#2800): remove the regex length ceiling and count only prose mentions Review found two real defects in the docs parity gate. The never-throws contract was false: building a RegExp from a declared flag or section title throws SyntaxError past ~100k chars, and Phase 2 admits overlay lanes whose declared strings are untrusted in length. Every one of these matches is literal, so String.includes replaces the regex outright, which also deletes escapeLiteral and the llama.cpp escaping it existed for. Arm 1 was context-blind: a flag mentioned only inside a fenced example or a commented-out row counted as documented. Both are stripped before matching. Also advertises all 13 lane flags in the argument-hint and corrects a stale eleven-lane count in the slug grammar note. * test(#2800): repoint the convergence suite off deleted workflow text The derived flag loop deleted the literal per-flag grep lines four tests matched on. Two of those failed loudly. The behavioral and property tests failed SILENTLY instead: their end marker no longer resolved, so the parse block extracted empty and both passed vacuously, and the property test's gsd_run stub had a no-op default that hid it. All now share one extractor and execute the real deployed block through a gsd_run shim backed by the actual binary. The whitelist assertions become an anti-parity check: re-adding a hand-written flag list must fail. Also repairs two vacuous cases in the docs parity suite. The unreadable-doc test called its own mock rather than the reader, and the integration test bounded nothing, so a doc losing its marker would have been silently skipped and still passed green. * fix(#2800): run the derived flag loop after the launcher preamble The remote matrix caught a real runtime bug, not a test artifact. In autonomous.md and plan-review-convergence.md the launcher preamble that defines gsd_run lives in a separate, LATER bash fence than the derived loop. Each fence is its own shell, so gsd_run was undefined where the loop ran: the command substitution yielded nothing and zero reviewer flags would have been forwarded. Worse than the drift this epic fixes, and silent. The whole CONVERGENCE_ARGS construction moves as one unit, because the --max-cycles append sits between the loop and the preamble and would otherwise have run against an uninitialized variable and then been dropped by the relocated initializer. Also documents all 13 lane flags in help/modes/full.md, which the repo gates bidirectionally against each command's argument-hint. * test(#2800): repoint the two converge suites off deleted flag literals Both asserted workflow.includes('--codex') against the hand-enumerated list the derived loop removed. They now assert the derivation itself, keep --all and --text (convergence controls, still literal), and add an anti-parity guard so re-adding a hardcoded list fails. The lost pass-through proof is replaced with a real one: every flag the tests used to hardcode is asserted present in the actual roster emitted by the binary, which is the property the old assertion was protecting. * test(#2800): acknowledge the workflow byte growth from the derived flag loop * chore(#2800): backfill changeset pr number to 2882 * fix(#2800): strip HTML comments to a fixed point in the parity gate CodeQL js/incomplete-multi-character-sanitization (high) on PR #2882: the single-pass <!--...--> strip can leave a live <!-- behind, so a join-trick construction smuggles a commented-out row past the gate and it counts as documented. Not an injection risk here since nothing is rendered, but it is the exact false pass this helper exists to prevent. Strips to a fixed point, then treats any surviving opener as unterminated so the multi-line branch closes it on a later line. Terminates because every pass strictly shortens the string. * test(#2800): pin the comment-smuggling regression with a real reproducer The obvious fixture for this class does not reproduce it: <!--<!---->--> leaves a dangling --> rather than a live <!--, and is caught either way, so it would have passed with and without the fix. The join-trick construction (<!- + <!--DUMMY--> + -...-->), the <scr<script>ipt> shape, genuinely regresses on the single-pass strip and is what the test now uses. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
b7b5c3712c |
fix(#2762): chunked --reviews replans instead of no-op + outline resume marker written to file (#2887)
* test(#2762): chunked --reviews must replan, not no-op (outline marker + per-plan --reviews exception) * fix(#2762): chunked --reviews replans plans instead of skipping 100% + outline resume marker written to file Defect A: §8.5.1 outline resume-check greped for a marker the agent only RETURNED (never wrote to the file) → outline always re-ran (broke crash-resume). Fix: the outline agent writes ## OUTLINE COMPLETE into the file. Defect B: §8.5.2 per-plan resume-check skipped any plan with frontmatter, no --reviews exception → --reviews skipped 100% of plans (contradicted §6 'go straight to replanning'). Fix: gate the skip on --reviews being ABSENT. Crash-resume (non-reviews) still skips. Condensed adjacent §8.5 prose to keep plan-phase.md under the 94519B cap (net -33B). * chore(#2762): changeset fragment * chore(#2762): backfill changeset PR number (2887) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
3af1941948 |
fix(#2772): resolve four discuss-phase text inconsistencies (dead MAX_PASSES read, gate-prompts drift, circular auto_advance, answer_validation drift) (#2886)
* test(#2772): structural guards for the four discuss-phase text inconsistencies * fix(#2772): resolve four discuss-phase text inconsistencies 1. auto.md: remove the dead MAX_PASSES/max_discuss_passes config read (contradicted the mandated single-pass rule + wasted a shim invocation per auto run). 2. gate-prompts.md: context-handling options now match the actual check_existing flow (Update it | View it | Skip, not Overwrite|Append|Cancel); gray-area-option no longer mandates 'Let Claude decide' (contradicts discuss-phase.md's no-cop-out rule). 3. discuss-phase.md: auto_advance fallback ends the workflow instead of routing back to the already-run confirm_creation step (circular). 4. discuss-phase-assumptions.md: re-sync answer_validation to the parent canonical block (had drifted — lost the 'Other' empty-text branch). * chore(#2772): changeset fragment * fix+test(#2772): also fix the assumptions auto_advance circularity (review minor 1) + add positive test anchors (review minor 2) The sibling discuss-phase-assumptions.md had the identical auto_advance→confirm_creation circularity; fix it the same way (end the workflow). Add positive anchors to both auto_advance tests so a re-phrased regression can't slip past. File #2885 for the dead max_discuss_passes config still advertised in settings/registry/docs (review minor 3). * fix(#2772): keep discuss-phase.md under the 32000B #717 cap + ack assumptions growth The auto_advance fixes + the assumptions answer_validation re-sync grew both files past the emitted-attribution gate (and discuss-phase.md past the #717 32000B cap). Condense the auto_advance prose in both files (discuss-phase.md now net -11, under cap; auto.md already net -4650 from the MAX_PASSES shim removal). Add discuss-phase-assumptions.md to tests/emitted-drift-ack.json for its residual +220 (answer_validation re-sync + auto_advance fix). * chore(#2772): backfill changeset PR number (2886) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
dbb0a653be |
fix(#2771): advisor mode spawns registered gsd-advisor-researcher subagent instead of general-purpose (#2884)
* test(#2771): advisor mode must spawn gsd-advisor-researcher, not general-purpose * fix(#2771): spawn registered gsd-advisor-researcher subagent instead of general-purpose in advisor mode universal-anti-patterns rule 10 (injected into discuss-phase via <required_reading>) says NEVER use non-GSD agent types. The advisor mode spawned general-purpose and manually told the agent to read the def — but gsd-advisor-researcher IS registered, so spawning by type auto-loads it. Drop the manual-read prompt line (re-specifying the def is a drift risk) and use the registered type. * chore(#2771): changeset fragment (mentions follow-up #2883) * test(#2771): widen manual-read-line regex to deny phrasing variants (review minor) /read\s+@.*gsd-advisor-researcher\.md/i (case-insensitive, any 'read @' lead-in) so a drift variant like 'Read @' or 'Load @' can't sneak the manual-def-read back in. * chore(#2771): backfill changeset PR number (2884) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
185da024cb |
fix(#2770): decision-coverage gate fails closed on empty arg + workflow recomputes CONTEXT_PATH in-block (#2881)
* test(#2770): empty contextPath argument must fail closed, not green-skip the decision-coverage gate The handler conflated empty-arg (caller error) with file-missing (legitimate skip), returning passed:true/skipped on an empty argument. Add: empty arg → passed:false; real-path-to-absent-file → legitimate green skip preserved; omitted arg → fail closed. * fix(#2770): decision-coverage gate fails closed on empty arg + workflow recomputes CONTEXT_PATH in-block Handler (check-command-router.cts): split the guard — empty/missing contextPath argument is a caller error (fail closed, passed:false, mirrors #1365); a real path whose file genuinely does not exist keeps the legitimate green skip. Workflow (plan-phase.md): recompute CONTEXT_PATH inside the consuming Bash block (it was set in the step-1 init block, which does not survive into the separately- spawned gate block — so the gate ran with an empty arg and silently green-skipped). * chore(#2770): changeset fragment * fix(#2770): guard workflow empty-glob case (review blocker) + update drift-guard test The handler now fails closed on an empty contextPath arg, so the workflow's unguarded glob (empty when a phase genuinely has no CONTEXT.md) would invoke the gate with an empty arg → passed:false → exit 1, hard-halting the legitimate 'Continue without context' plan-phase path. Guard the empty-glob case: only run the gate when a CONTEXT.md actually exists. Update the F1 drift-guard test (which gave false coverage — it only checked for the ${CONTEXT_PATH} token) to assert the in-block recompute AND the empty-glob guard. * fix(#2770): keep plan-phase.md under ADR-857 size cap + ack emitted drift + fix drift-guard window The workflow fix grew plan-phase.md past the ADR-857 phase-6 size cap (94519B) and triggered emitted-attribution. Condense adjacent §13a prose/JSON to offset (net +89B, under cap). Add tests/emitted-drift-ack.json acknowledging the residual growth. Widen the drift-guard test window (the gate invocation is now nested in the empty-glob guard, so the old 400-char window missed the glob recompute). * chore(#2770): backfill changeset PR number (2881) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
7e9ce08e2c |
test(#2736): property-test parsePhaseFromProse name precedence (#2878)
#2821 reworked parsePhaseFromProse's NAME precedence (dash-vs-paren choice, status-keyword tails, `Milestone:` tails, paren-stripped separator search, the lone-ALL-CAPS-tail rule) but shipped it pinned only by hand-picked examples. The two pre-existing property blocks cover phase-token ANCHORING (#2111) and "N of M" phase extraction; neither touches name precedence. This adds nine properties over the canonical parser. P1 and P9 are the delta guards: both fail against the pre-#2821 paren-first parser (proved with a standalone mutation harness), because each requires a genuine em-dash name to win over a co-present parenthetical. P9 additionally exercises the paren-stripped separator search, since the losing parenthetical itself contains an em-dash. P2-P4 are characterization tests for precedence contracts both parser versions satisfy; P5-P8 pin totality and phase-token extraction, which #2821 left alone. The section comment states which is which so a future reader does not mistake the characterization tests for delta guards. The generator's status-word exclusion filter is a test-local mirror of the private, unexported STATUSY_TAIL_RE. A divergence-guard test pins that mirror to observable parser behavior (not source text, which the no-source-grep rule forbids), so an implementation vocabulary change fails loudly instead of silently weakening every property that depends on the filter. Also clears six pre-existing no-unused-vars lint warnings surfaced by the lint run for this change (dead bindings in four unrelated test files, deleted rather than underscore-renamed); removing the never-called openCodeBlock helper cascaded to its now-unused fs/path/reviewPath consts in both fold scopes. Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
7cbfb2819f |
fix(#2764): extend no-path-literal-in-assert to membership/substring checks (catches the #2728 Windows CI shape) (#2879)
* test(#2764): no-path-literal-in-assert must flag membership/substring checks over path-returners The equality-only visitor missed .includes/.indexOf/.startsWith/.endsWith over a path-returning receiver (directly or via a .map() hop) — these pass lint and fail on Windows. Add RuleTester invalid cases for each shape + valid cases for the suppressions (POSIX-normalized receiver, non-path receiver, no-slash arg). * fix(#2764): extend no-path-literal-in-assert to membership/substring checks over path-returners Add a third CallExpression shape: .includes/.indexOf/.startsWith/.endsWith/.match over a receiver that traces to a path-returning call (directly, or through ONE .map(f => path-…) hop — the #2728 shape that passed lint and failed Windows CI). Reuse isPathReturningCall/isPosixSlashStringLiteral/isPosixNormalizerCall; respect the normalizer + Windows-excluded suppressions. The rule is scoped to tests/**/*.test.cjs. * chore(#2764): changeset fragment * test+docs(#2764): add .match test coverage, Windows-excluded membership case, fix doc accuracy (review findings) - MAJOR: .match was supported but untested — add invalid (.match with slash) + valid (.match no-slash) RuleTester cases. - MINOR: Windows-excluded membership case was claimed by the matrix but absent — add a platform-guard valid case in membership form. - MINOR: .map() hop comment said 'ONE' but recursion allows nesting — reword to 'recursively'. - Add known-boundary (d) note for the membership/.match shape. - Fix a premature comment-close from a literal */ in the glob example. * chore(#2764): backfill changeset PR number (2879) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
ef7a71fc09 |
fix(#2754): make parseStateMd CRLF-safe — frontmatter no longer drops on Windows-authored STATE.md (#2865)
* test(#2754): parseStateMd must parse CRLF STATE.md identically to LF The frontmatter fence regex and downstream splits used literal \n, so a CRLF STATE.md dropped the ENTIRE frontmatter block. Assert CRLF/LF parity (the production contract) across full frontmatter, next_phases flow + block forms, the progress nested block, and null handling. * fix(#2754): make parseStateMd CRLF-safe — frontmatter fence + splits use \r?\n The fence regex, scalar-line split, next_phases block-list regex, and progress block regex all used literal \n, so a CRLF (Windows-authored) STATE.md dropped the ENTIRE frontmatter block — every statusline field was silently absent. Use \r?\n throughout, mirroring the CRLF-safe extractFrontmatter in src/frontmatter.cts. LF behavior unchanged. * chore(#2754): changeset fragment * test(#2754): pin parseStateMd↔extractFrontmatter parity (Generative-Fix Divergence guard) The statusline parseStateMd and the canonical extractFrontmatter both derive GSD state from STATE.md frontmatter and diverged once already (the CRLF bug this PR fixes). Add a cross-parser parity assertion (CLAUDE.md parallel-surfaces rule) over the overlapping fields, under both LF and CRLF, so a future divergence on a scalar shape or line ending is caught here. (isolated-review minor finding) * chore(#2754): backfill changeset PR number (2865) --------- Co-authored-by: Test <test@example.com> |
||
|
0xdhx
|
82ca13f5a5 |
fix(#2736): write current_phase_name from the transition intent, not the lossy prose round-trip (#2821)
* fix(#2736): intent-first current_phase_name on transitions; dash-first prose precedence Primary: completePhase (adapter) and beginPhase (via readModifyWriteStateMd options) pass the intent-held display name to syncStateFrontmatter as an authoritative override, applied after every derive/preserve/carry-forward step — so the lossy prose round-trip can never destroy a name the transition just resolved. Names containing a parenthetical (`Closer-ruling measurement (D1a)`) now land in frontmatter verbatim instead of collapsing to the parenthetical (`D1a`). Secondary (#1695 AC #3 residual): parsePhaseFromProse prefers the em-dash name when it is a genuine name (not a status keyword, not a `Milestone:` tail), else falls back to the parenthetical — satisfying both first-party writer shapes (`N — Name (aside)` and `N (Name) — EXECUTING`). Still lossy for paren-containing names, which is why the intent-first override is the primary fix. plannedPhase carries no name in its intent, so it is naturally out of scope. Fixes #2736 * docs(changeset): backfill PR number for #2736 fragment * fix(#2736): drop an unnecessary type assertion on result.data StateTransitionResult.data is already `Record<string, unknown> | undefined`, so the cast was a no-op and tripped @typescript-eslint/no-unnecessary-type- assertion (CI lint-tests red on the first push; every test lane was green). --------- Co-authored-by: Tom Boucher <trekkie@nomorestars.com> |
||
|
Tom Boucher
|
3f6b063fbb |
chore(#2799): invoke_reviewers and write_reviews iterate declared lanes (#2861)
* chore(#2799): resolve reviewer lanes into executable invocation plans Phase 5b of ADR-2782. Adds the resolver and runner that let invoke_reviewers iterate declared lanes instead of hand-authored per-CLI bash. Five additive descriptor amendments, each forced by a lane that ships today: - LaneHandler gains 'opencode' — the lane rebuilds its review from assistant text parts of a --format json stream; a plain stdout copy re-breaks #1936. - modelConfigKey — antigravity's key is review.models.agy, not .antigravity, so resolving by slug silently dropped a configured model. - defaultHost/fallbackModel — Phase 4 federated every *_host with a default of empty string; the real fallback only existed in the bash. - args becomes an argv template with a closed four-placeholder vocabulary. Positional splicing produced 'codex --model M -o F exec --ephemeral', which is not a valid invocation: codex injects in the middle, twice. - kimi-code lane, with the bounded command-capability probe (needle --output-format) that tells Kimi Code from the legacy python kimi-cli. Parity gate re-pointed: the workflow-text families it scanned are the text this phase deletes, so they are replaced by descriptor-to-registry parity plus an anti-parity check that no bespoke leg returns. jq, curl and external timeout/gtimeout all drop out of the review path. Refs #2782 * chore(#2799): add review-lane query surface and widen the manifest vocabulary Adds the gsd-tools 'review-lane' route (plan/invoke/sections) the workflow loops over, projects all twelve lanes into their capability manifests, and widens capability-validator for the amendments. opencode admitted to VALID_LANE_HANDLERS under the second arm of the enum's own admission rule: one lane, justified by a documented upstream defect data cannot express (#1936 — the agent can end its turn with zero output tokens and --format default then drops the assistant text entirely). Two bugs caught by an end-to-end stub run and fixed here: - loadConfigResolved returns a provenance wrapper, not the config; using it directly resolved every key to undefined, which reads as 'nothing configured' and silently dropped every model override. - hasBinary used shell:true with an args array (Node 26 DEP0190). Replaced with a PATH scan that spawns nothing at all. Refs #2782 * chore(#2799): iterate declared lanes in invoke_reviewers and write_reviews Replaces the eleven hand-authored per-CLI bash legs with a loop over resolved lanes, and renders REVIEWS.md sections from each lane's declared reviewsSection instead of thirteen hardcoded headings. review.md drops from 1104 lines to 507 (61KB to 28.7KB). Parity gate re-pointed, as agreed: the leg-marker and section-heading families scanned exactly the text this phase deletes, so they are replaced by descriptor-to-registry parity in both directions, plus an anti-parity check that fires if a bespoke leg is ever re-added. Enum, emitting sites and the Object.keys lock moved together. The budget-trim helper is hoisted out of the Ollama leg: it was always lane-agnostic, and any lane may now declare a promptBudgetKey. Refs #2782 * feat(#2799): bind the consented egress host and re-verify it at invocation Completes ADR-2782 D5. Rule 1 was recorded in the ADR as delivered by Phase 3 but was not implemented: ConsentRecord had no host field and nothing in the tree bound one, so this phase's rule-4 comparison had no baseline. ConsentRecord gains an OPTIONAL reviewerHost. Optional is the whole design: isValidConsentRecord does not require it, so every record already on disk stays valid and no re-consent storm fires (D4 rule 5). It is deliberately excluded from disclosureSignature — the loader has no config resolver, so folding a config-derived value in would make loader and lifecycle compute different signatures for the same manifest and re-prompt forever. Install resolves hostConfigKey (falling back to the lane's declared defaultHost, which is what the invocation path uses) and records it. Invocation re-resolves and blocks on mismatch rather than silently redirecting. Absence allows: no record, or a record predating the field, means nothing to compare — denying there would break every existing local-model user on upgrade. Refs #2782 * test(#2799): cover the resolver, runner and handlers; retarget the parity suites Adds the golden invocation-plan table (one row per shipped lane, derived from the bash legs rather than the descriptor types) plus runner coverage for the probe, empty-output policy, the three handlers and the egress check. Retargets the existing suites onto the new contract: descriptor-to-registry parity, the anti-parity check, the opencode handler, and the twelfth lane. Two corrections found by running them: - modelConfigKey was required; that breaks D4 rule 2, since a reviewer manifest authored before this phase would fail validation on upgrade. It is optional, read as null when absent. - the antigravity non-zero-exit test pre-seeded the transcript, which asserted that a STALE entry leaks through — the exact bug the watermark prevents. The spawn now appends, as the real tool does. Refs #2782 * fix(#2799): restore agy --add-dir and the self-report prompt in the handler Retargeting the three legacy reviewer suites off the deleted bash surfaced two real regressions in the port, both #2176: - --add-dir was dropped. Without it agy's permission context never receives the cwd repo, so the agent anchors on its own scratch dir and reviews the plan text in isolation — the exact failure the Review Instructions forbid. It is capability-probed, because an older agy rejects the unknown flag outright and a lane that fails to start is worse than one running on the prompt anchor. - the prompt lost the clause mandating a REVIEWED-WITHOUT-REPO-ACCESS self-report, which is what makes a blind review distinguishable from a grounded one. antigravity now builds its own prompt variant. Also ports the #2073 mode-2 cli.log diagnostic, which was dropped: a pinned model that 404s exits 0 with empty stdout AND an empty transcript, so agy's own log is the only evidence that anything failed. The three suites now assert against the plan and the handler instead of matching fence text, so they no longer need allow-test-rule exemptions. Refs #2782 * docs(#2799): document the declared lanes, the new flag, and dropped prerequisites COMMANDS.md gains --kimi-code and replaces the jq-prerequisite paragraph, which is now false: no lane requires jq, curl or an external timeout. Adds the changed-egress-destination behavior, since a blocked lane is something a user can hit. CONFIGURATION.md records that the model config key is declared per lane rather than derived from the flag — antigravity's is review.models.agy — and adds review.models.kimi-code. reviewer-instances.md now routes an instance through its lane's single invocation seam instead of a copied per-adapter bash block, which is what lets a cross-cutting fix reach instances for free. That required implementing the --model/--agent/--as flags it documents; --model re-resolves through the lane's argv template rather than splicing, so the flag lands where the lane declares it rather than ahead of a subcommand. CONTEXT.md glossary gains both new modules. Refs #2782 * chore(#2799): drop the stale emitted-drift acknowledgment The only entry was #2797's, acknowledging COMMENT-ONLY GROWTH in review.md. That file now shrinks by ~32KB and every emitted hash that moved is attributable to this diff, so the ack no longer explains anything. Removing the last entry means removing the file: its presence is the alarm, and an empty one signals nothing. Verified by deleting it and re-running the attribution and provenance gates plus lint:ci — all green without it. Refs #2782 * docs(#2799): record the Phase 5b vocabulary widenings in ADR-2782 Five additive amendments, each forced by a lane that ships today, plus two corrections the phase had to make rather than work around: - D5 rule 1 was recorded as delivered by Phase 3 and was not implemented, so this phase's rule-4 comparison had no baseline. Recorded because an ADR asserting a rule was delivered is exactly what stops a later phase checking. - The DEFECT.GENERATIVE-FIX gate is re-pointed: its workflow-text families scanned the text this phase deletes. Also records that D7's 'skip the probe where no bounding mechanism exists' carve-out is obsolete — in practice it meant the Antigravity lane ran unbounded on every stock macOS host, which ships neither timeout nor gtimeout. Refs #2782 * fix(#2799): close four defects found by adversarial review Two confirmed bugs, both reproduced before fixing: - resolveLanePlan was not total. An openai-http lane with a missing or non-object invoke dereferenced inv.hostConfigKey and threw, contradicting the module's own documented contract; the spawn branch guarded correctly and the http branch did not. The CLI seam resolves every selected lane in one map, so one malformed overlay manifest would have aborted the whole review rather than dropping its own lane. Guarded, plus a per-lane try/catch at the seam so a throw can never take down siblings. - A reviewer-instance model was silently dropped for any lane declaring modelConfigKey null (cursor, qwen, coderabbit). reviewer_instances validates that cli is a known slug but never that the slug accepts a model, so a user could configure one, get a clean run, and never learn a different model reviewed their plan. Now warns explicitly. Two hardening fixes: - The slug is concatenated into artifact paths, so LANE_SLUG_RE is enforced in the resolver rather than inherited from a validator that does not run on this path — the module documents itself as the overlay-manifest trust boundary, so it should not depend on someone else having checked. - normalizeHost mangled a scheme-less value: new URL('localhost:11434') parses with an empty hostname, so it became 'localhost://11434' and was compared and requested as if real. An empty hostname now means not-a-URL. Also documents the one gap that cannot be closed here: the antigravity watermark is keyed by workspace, so two concurrent reviews of the same repo share a transcript. agy exposes no per-invocation id to filter on, so the handler now states which half of its never-stale guarantee actually holds. Refs #2782 * test(#2799): retarget the remaining eight review.md-asserting suites The remote runner found 37 failures the local sweep missed (it hit the shell's two-minute cap before reaching these). All eight extract per-CLI bash from review.md that this phase deletes; each protects a real invariant, so each is retargeted onto the plan, the runner or the handler rather than removed. Three real defects surfaced by doing so: - effort args never reached ANY lane. model-resolver.cjs exports no resolveExecution, so effortFor silently returned [] every time. Restored by calling the same bounded resolve-execution query the bash legs used — and NOT with --raw, which prints the resolved effort rather than the picked field, so claude got 'low' instead of '--effort low'. - the timeout guidance lost 'a silent empty output is a timeout kill, not a crash' — the operator note that exists because of the Codex 0xc0000142 misdiagnosis. Restored. - the opencode handler dropped EMPTY assistant text parts. The shipped jq was , and only substitutes for false/null — an empty string is truthy in jq and contributed a blank line. Found by a property test shrinking to ['', '']. The opencode property suite no longer spawns jq at all, which deletes the #2099 hang mechanism it was architected around rather than mitigating it. Refs #2782 * fix(#2799): register the two new generated modules, and untrack them The remote runner caught build output committed to git. Both new modules compile from src/*.cts into gsd-core/bin/lib/*.cjs, and every sibling generated that way is gitignored and eslint-ignored (ADR-457) - including Phase 1's own review-lane-descriptor.cjs. Mine were neither, so repo-invariants' "each bin/lib/*.cjs is linted xor ignored according to migration state" failed. Registered both in .gitignore and eslint.config.mjs alongside the Phase 1 module, and dropped them from the index. Nothing about the shipped behaviour changes; the artifacts are rebuilt by build:lib. This is the new-.cts-module registration ripple, and it is the one part of it I had not completed - the CONTEXT.md glossary and the inventory manifest were already done. Refs #2782 * chore(#2799): backfill changeset pr number to 2861 * chore(#2799): backfill changeset pr number to 2861 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
97e9fd50bd |
fix(#2752): make tool_input.path authoritative over model-controlled file_path in gsd-phase-boundary.sh (#2860)
* test(#2752): path is authoritative over model-controlled file_path in phase-boundary hook Rewrite the #2304 precedence test (which pinned the buggy file_path-wins behavior with an impossible no-tool_name {file_path,path} payload) to assert the correct precedence with realistic Kimi-shaped payloads: a real .planning/ write with a decoy file_path must still fire the reminder (suppression repro), and a write elsewhere with a decoy .planning/ file_path must NOT fabricate one (fabrication repro). Update the parity vocabulary alarm to the new expression. * fix(#2752): make tool_input.path authoritative over model-controlled file_path in gsd-phase-boundary.sh The hook consulted file_path first, path second. kimi-cli executes on path and sends path only; file_path on a Kimi payload is always model-supplied. So a model-supplied decoy file_path could suppress the reminder for a real .planning/ write or fabricate one for a file never touched. Flip the precedence so path is authoritative and file_path is the fallback (Claude Code emits file_path and no path, so the fallback must remain). Mirrors the #2595 JS-guard fix. * chore(#2752): changeset fragment * chore(#2752): clarify comment/changeset — JS guards use upstream normalization, shell hook applies precedence directly (review minor 1) * chore(#2752): backfill changeset PR number (2860) --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
aa19e3478c |
fix(#2854): pin the emitted gate to the base the tree was merged with (#2859)
* test(#2854): failing-first coverage for CI baseline export provenance Extracts the export decision out of main() behind injected IO so it is unit-testable, preserving today's export-whenever-present semantics, and adds the matrix that proves those semantics are wrong. The PR lane restores the emitted baseline keyed on the PR's recorded base sha while the gate resolves the base ref live, so the two drift whenever next advances mid-flight. The restore was published straight to GSD_EMITTED_BASELINE, where a mismatch is fatal, turning a recoverable cache into a hard failure on diffs that touched nothing related. Also renames the stale-env fixture from 'from-cache-restore.json' to an operator-pin name: that fixture asserted the exact conflation this bug is, documenting the defect as intended behavior. Refs #2854 * fix(#2854): validate a restored baseline before publishing it as an operator pin GSD_EMITTED_BASELINE is an operator pin: resolveBaseline() treats a mismatch there as a hard stop, because the operator said "use this one". CI published its cache restore to that same variable whenever the file merely existed, so a restore keyed on the PR's recorded base sha - while the gate resolves the base ref live - turned a recoverable cache into a fatal error whenever next advanced mid-flight. Required tests went red on diffs that touched nothing related, and named a test file the contributor never opened. The export step is the boundary, so it is the boundary that validates. It now publishes only a baseline already valid for the sha under test, judged by validateBaseline so the staleness rule keeps one definition. Anything refused is left to be found via the cache path, where a mismatch degrades to the in-job build exactly as ADR-2719 SS5 specifies. The operator hard stop is untouched, and the fast path still hits on a current cache. Also reports the sources actually reached rather than asserting all three ran: the failure message claimed an in-job build it had returned before calling, sending contributors after a rebuild that never happened. Fixes #2854 * fix(#2854): pin the emitted gate to the base the tree was actually merged with The differential compared a tree built on one commit against a baseline at a different one. "Rebase check" merges pull_request.base.sha, pinned by #2472 so all 12 matrix jobs agree on one tree, but resolveBase() fell through to origin/next, which fetch-depth 0 leaves at the live tip. Nothing set GSD_EMITTED_BASE, so whenever next advanced mid-flight the two disagreed. The cached baseline, keyed on base.sha, was correct for that tree and was rejected as STALE by a target that was not. Required tests went red on diffs that touched nothing related, naming a test file the contributor never opened. The near miss is the worse half: had resolution gotten past the baseline step, a baseline at the live tip would have attributed commits merged to next in between to the PR under test. The hard stop was shielding us from a wrong answer, so making it fall through would have made this worse. Pins GSD_EMITTED_BASE to the same expression as CI_REBASE_BASE_SHA in every rebase-merged job, with a parity test asserting the two cannot diverge. That parity check immediately caught a third lane, test-inert, that merges a pinned base and had been missed. Fixes #2854 * fix(#2854): keep the export step self-contained across the package boundary scripts/ ships in the npm tarball and tests/ does not, so requiring the validator across that boundary is MODULE_NOT_FOUND in a published install. The export step now reads the same GSD_EMITTED_BASE pin the gate resolves through, and applies a cheap self-contained precondition; validateBaseline remains the sole authority and still runs downstream on whatever is published, so there is no second opinion to drift. Reading the pin rather than re-deriving a base is the point: a second, divergent base lookup is exactly what caused this bug. The same hazard pre-exists in scripts/gen-emitted-baseline.cjs, which ships and requires three tests/ modules. Filed as #2858 rather than folded in: fixing it means relocating the shared helpers out of tests/ and updating every consumer, which would bury this change. Refs #2854 * fix(#2854): stop announcing the restored cache through the operator-pin door Two independent reviewers found the same blocker in the previous approach. Validating before publishing to GSD_EMITTED_BASELINE only narrowed the hole: the precondition gated on sha equality alone, so a document with a MATCHING sha but a wrong schema version or malformed manifests was still announced as an operator pin and still hard-stopped downstream. That reproduces this bug's own class, triggered by malformation instead of staleness. The step was never load-bearing. The cache restores to .gsd-cache/emitted-baseline.json, which is resolveBaseline's DEFAULT_CACHE_PATH and is read whether or not anything announces it. Publishing the same file to the pin door could only ever convert recoverable into fatal, so the step and its script are deleted rather than made cleverer. Every failure mode now degrades to the in-job build by construction, and validateBaseline is once again the only thing that judges a baseline. Coverage moves to where the behaviour lives: stale sha, wrong schema version, manifests array/absent, non-object documents, unreadable file, and the 39/40/41 hex boundary all assert degradation via the cache path. Adds the empty-pin case a reviewer flagged as untested - the pin is job-level env, so on push events it renders as an empty string, and only baseRefCandidates' truthy check keeps it out of the candidate list. Fixes #2854 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
b36e3b7e1f |
fix(#2751): normalize bare gsd-tools command-position calls to gsd_run in shipped source (#2851)
* test(#2751): regression guard — no command-position bare gsd-tools calls Agents/workflows instructed bare `gsd-tools <verb>` invocations that fail with 'command not found' on a shim-only install (#725 fixed only the Codex conversion pipeline; the Claude-facing source shipped them verbatim). Adds a source-text guard (allow-test-rule: source-text-is-the-product) scanning agents/*.md + gsd-core/workflows/*.md for the operative shape `gsd-tools <verb> <arg>`, excluding command -v probes / resolver definitions, with a documented PROSE_ALLOWLIST for descriptive mentions that name the command without instructing literal invocation. A stale-allowlist check ensures entries stay real. RED first; source fix lands next commit. * fix(#2751): normalize bare gsd-tools calls to gsd_run in Claude-facing source The 12 command-position bare `gsd-tools <verb>` instructions across agents/ and gsd-core/workflows/ failed with 'command not found' on a shim-only install (no gsd-tools binary on PATH). #725 fixed this only for the Codex install-conversion pipeline; the Claude-facing SOURCE shipped the bare calls verbatim, and new ones kept accumulating (new-project.md:114 landed 12 days AFTER #725 closed). Rewrite each operative site to the portable `gsd_run` resolver that the same files already define (3-18x each) — a pure command-position token swap preserving all arguments, flags, --files, and surrounding prose. Every runtime now benefits from one source change instead of each needing its own converter. Touched sites (12): gsd-intel-updater (validate/snapshot/extract-exports), gsd-code-fixer (query commit), gsd-planner (learnings.query), gsd-project- researcher (websearch/research-plan/classify-confidence), gsd-phase-researcher (websearch/research-plan/classify-confidence), new-project (project-instruction- file / commit --files), new-milestone (commit --files). Preserves command -v gsd-tools probes, resolver-snippet definitions, and the 4 descriptive prose mentions that NAME the command without instructing invocation. RED @ 1bb12ba1. * chore(#2751): allow-test-rule issue ref + changeset fragment Add the (#2751) tracking ref to the source-text-is-the-product annotation per ADR-456, and the .changeset Fixed fragment (pr:0, backfilled post-PR). * fix(#2751): convert remaining command-position bare gsd-tools calls (verify-summary, windows, worktree, smart-entry, quick-tasks-append) Isolated adversarial review (Step 4) found the first pass missed genuine command-position bare calls because the regression test's hand-maintained 6-verb list silently false-passed verify-summary (the 'verify' branch matched the prefix then died on the hyphen) and omitted windows/worktree/smart-entry/ quick-tasks-append entirely. Convert these 8 additional operative sites across new-project.md, new-milestone.md, ship.md, execute-phase.md, progress.md, smart-entry.md, quick.md. * chore(#2751): backfill changeset PR number (2851) * fix(#2751): normalize allowlist paths to forward slashes — Windows path-separator false-flag The PROSE_ALLOWLIST is keyed by file:line using forward-slash paths, but path.relative() returns backslash separators on Windows, so the allowlist lookup failed and the 6 descriptive mentions were flagged as offenders on the windows-latest CI lane. Normalize rel to forward slashes before the lookup so the allowlist matches identically on every OS. --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
0408276791 |
chore(#2797): federate reviewer config keys off the central schema (#2841)
* chore(#2797): federate reviewer config keys off the central schema Phase 4 of epic #2782 (ADR-2782 D9, config half). Runs AFTER 5a per the ADR's swap amendment: a federated config slice lives inside a capabilities/<id>/capability.json, and three of the five key families had no capability directory until 5a created them. Four key families move to the lanes that use them; the central-schema removal and the federated addition land in this one commit because the exclusivity invariant fails the build on a key present in both. review.max_prompt_tokens, review.default_reviewers and review.reviewer_instances describe policy ACROSS lanes and stay central. Two things the issue did not name, both found while building it: 1. THE EXCLUSIVITY GATE WAS BLIND TO PATTERNS. It compared federated keys against manifest.validKeys only, and two of the four families (review.models.<slug>, review.max_prompt_tokens_per_reviewer.<slug>) were pattern-backed. That is not cosmetic: isCentralConfigKey consults those patterns and mergeFederatedConfig skips every key for which it returns true, so declaring a slice while the pattern survived would have shipped an INERT slice behind a green gate — the exact half-migrated shape the invariant exists to prevent. The gate now loads the patterns from the same manifest the runtime reads. 2. AN UNSET PER-LANE BUDGET NOW RESOLVES TO 0, NOT NOT-FOUND, because a federated key always resolves to its declared default. The three fallback guards in review.md checked only empty-or-"null", so a user who set the GLOBAL review.max_prompt_tokens would have silently lost trimming on the HTTP lanes. The guards now treat 0 as unset. D9 says review.models.<slug> is owned by "the lane whose slug it names". That is false for one lane: the shipped key is review.models.agy while the slug is antigravity. Ownership follows the lane; the key name is preserved, because renaming would break every config that sets it. Existing tests updated rather than left asserting the old world: config-get on a cleared federated key yields empty instead of not-found (what #2046 actually protects — never persisting the literal "null" — is unchanged and still asserted); the config-schema dynamic pattern representative moves to reviewer_instances; the prototype-pollution guard case moves to a surviving dynamic prefix so alert #26 keeps its coverage, with a new case asserting the old key is now rejected earlier; and Phase 2's harvest-widening inertness assertion becomes an ownership assertion, since Phase 4 is what consumes it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): use a -1 sentinel so an explicit per-lane budget of 0 survives A federated config key always resolves to its declared default, so an unset per-lane prompt budget needed a value the workflow could treat as 'not configured'. The first cut used 0 — which is wrong: 0 is already a LEGITIMATE per-lane budget meaning 'do not trim this lane' (the early-return guard in prepare_trimmed_prompt_for_reviewer). Treating it as unset would have silently switched a user who deliberately disabled trimming for one lane onto the global budget. The sentinel is now -1, which is not a valid token budget, so all three states stay distinguishable: unset falls back to global, an explicit 0 disables trimming for that lane, and an explicit N is used. Locked by three CLI round-trip tests. Surfaced by the isolated security reviewer before it crashed mid-run; verified independently against the shipped trim guard rather than taken on trust. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): update central-registration assertions and stay under the review.md cap The remote runner caught both; my local sweep missed the files. 1. tests/plan-review-convergence.test.cjs asserted the three local-server host keys are in VALID_CONFIG_KEYS. They are federated to their lane capabilities now, and the exclusivity invariant forbids a key living in both places. What #2306-local actually protects is that config-set ACCEPTS them, so that is what is asserted — via isValidConfigKey, the predicate config-set itself uses, which spans central and federated. A second assertion pins federated ownership, so a silent reversion back to the central schema fails too. 2. review.md exceeded the LARGE tier hard cap (62583 > 61440). That cap is a red line, not a budget to raise. The three per-lane budget guard comments were near-identical; condensed to one terse line each. 61371 bytes, 69 to spare. Real extraction to workflows/review/modes/ is Phase 5b/6 work. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): fail closed on a broken config-schema manifest; reconcile stale docs Isolated security review findings. MAJOR — loadCentralConfigPatterns failed OPEN. It swallowed a JSON parse error and returned [], while its sibling loadCentralConfigKeys, reading the SAME file, writes to stderr and throws ExitError(1) on that identical failure class. Fail-open here defeats the gate this function exists to feed: with zero patterns, validateCrossCapability's pattern-collision check silently passes and an inert federated slice ships green. It was masked in the one production call site only because loadCentralConfigKeys runs first against the same path — a coincidence of ordering, not a guarantee, and this function is exported and called standalone. The two now share a contract: ENOENT is the legitimate absent case, anything else throws loudly. A single unparseable PATTERN is still skipped, which degrades to "checked less" rather than blocking every build. The branch had zero coverage; it now has two tests (malformed JSON, EISDIR). MINOR — docs/CONFIGURATION.md still listed review.models.qwen and review.models.cursor as settable, ~770 lines below this PR's own new Ownership section. Those lanes take no model flag, so they declare no model key and config-set now rejects them. Rows removed; the missing review.models.agy row added; the per-reviewer budget row corrected to name only the lanes that own a budget key, and to document that a per-lane 0 disables trimming for that lane. Also fixes a shadowed "raw" binding introduced by the fail-closed change, which made the generator unrequirable — caught immediately by its own --check. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2797): backfill changeset pr number to 2841 * fix(#2452): make the base-ref mutation test hermetic against leaked GIT_* env tests/mutation-workflow-base-ref.test.cjs fails on PR branches while next stays green, and it is currently blocking at least three unrelated PRs (#2841, #2832, #2827) with: error: invalid object 100644 <sha> for 'base-N.txt' error: Error building trees The existing loop comment attributes this to `git add .` rehashing O(n^2) blobs "before the object write had landed" and works around it by staging one path per iteration. That is not the cause: sequential execFileSync calls cannot race each other's object writes, and the failure persisted after that change — it simply moved to a lower commit index. The cause is that the git() helper inherited the runner's environment. A leaked GIT_INDEX_FILE makes `git add` write into a DIFFERENT repository's index; GIT_OBJECT_DIRECTORY / GIT_ALTERNATE_OBJECT_DIRECTORIES send the blob to another object store; GIT_DIR / GIT_WORK_TREE redirect the whole operation. In every case `git commit` then cannot resolve a blob it just staged, which is precisely the error above. Verified by negative control: with GIT_DIR exported, this test fails on the unfixed helper (the git commands operate on the wrong repository entirely); with the helper stripping GIT_* it passes. The single-path staging is kept — it is genuinely less work — but it is no longer load bearing. Found while shipping #2797. Fixed in place rather than deferred: it is a defect surfaced during the work, and it is blocking other contributors. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2452): build the base-advance commits empty, removing the lost-object class The base-ref guard has been failing in CI with: error: invalid object 100644 <sha> for 'base-N.txt' error: Error building trees It is currently red on at least three unrelated PRs (#2841, #2832, #2827) while next stays green. Two theories have now been tried and neither held. #1881 blamed `git add .` rehashing O(n^2) blobs and switched to staging one path per iteration; the failure moved from commit 32 to commit 25 and carried on. The preceding commit here made the git helper hermetic against leaked GIT_* environment — that IS a real vulnerability (with GIT_DIR exported the helper operates on the wrong repository entirely, proven by negative control) but it produces a different error than CI reports, so it is not demonstrably the cause either. Neither trigger reproduces off-CI, so this stops guessing at the trigger and removes the failure CLASS instead. The loop needs base-branch DEPTH and nothing else: no assertion reads these commits' contents, and base-side files cannot appear in `origin/base...HEAD` regardless. `--allow-empty` writes no blob and no tree, so there is no object for the index to reference and lose. It is also far less work than 60 write+hash+index cycles. The guard still proves its mechanism: the test asserts that a --depth=1 base fetch FAILS and a full fetch resolves, so a broken topology would surface immediately rather than passing vacuously. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
4f6935e29b |
fix(#2717): write CommonJS marker for cursor/windsurf/codex staged .js hooks (#2846)
* test(#2717): CommonJS marker for cursor/windsurf/codex staged .js hooks Cursor/windsurf (skipSharedHooksInstall) and codex (!isCodex gate) stage .js hook scripts via dedicated paths that bypass installSharedHooksBundle — the only writer of the {"type":"commonjs"} marker. Under a config root declaring {"type":"module"}, Node loaded those scripts as ESM and every require() failed with 'require is not defined', silently disabling the runtime's hooks. Adds regression tests (RED first, fix lands next commit): - parametrized cursor/windsurf/codex install asserts hooks/package.json exists with exactly GSD's marker content; - end-to-end: a cursor require()-using hook loads under a planted ESM-typed config root without the require-is-not-defined error; - the ensureCommonJsMarker / removeCommonJsMarkerIfGsdOwned contract: GSD markers are removed on uninstall, user-authored package.json is never touched. * fix(#2717): write CommonJS marker for cursor/windsurf/codex staged .js hooks The {"type":"commonjs"} marker lived only inside installSharedHooksBundle, which cursor/windsurf (skipSharedHooksInstall) and codex (!isCodex gate) never reach. Their .js hooks are staged by dedicated paths, so under a config root declaring {"type":"module"} Node loaded them as ESM and every require() failed with 'require is not defined', silently disabling those runtimes' hooks. Decouple the marker write into a shared helper so any code path that stages .js hooks can ensure it lands in the SAME directory as the scripts: - src/runtime-hooks-surface.cts: add ensureCommonJsMarker(dir) + removeCommonJsMarkerIfGsdOwned(dir) (byte-identical content to installSharedHooksBundle's marker; preserves a user-authored package.json on both write and uninstall). Call ensureCommonJsMarker(hooksDir) from writeCursorHooksJson + writeWindsurfHooksJson; call removeCommonJsMarkerIfGsdOwned on their matching remove paths. Export both. - bin/install.js: call hooksSurface.ensureCommonJsMarker after the codex hook copy; call hooksSurface.removeCommonJsMarkerIfGsdOwned in the generic hooks-removal loop (safe no-op where no marker exists). No change to which runtimes receive the shared bundle, the !isCodex gate, skipSharedHooksInstall, or kimi/kimi-code/cline/copilot/trae/zcode (all unchanged — audit in the diagnosis). RED @ dbb7d2bb (6 failures: 3 missing markers + the ESM require error + missing helpers); GREEN pending. * docs(#2717): changeset fragment (pr:0, backfilled post-PR) * chore(#2717): regen codex/cursor/windsurf install-tree fixtures + attribution ack The fix adds hooks/package.json to those three runtimes' install trees (the new CommonJS marker), so the golden install-tree fixtures gain one path each (regenerated via npm run gen:install-tree). emitted-attribution (ADR-2719) flags the 3 emitted hooks/package.json paths under the hooks-built rule; acknowledge them. Also drops 5 spent ack entries left by now-merged PRs (#2694 code-review.md/code-review-fix.md, #2695 worker/registry, #2794 review.md) — they are stale on this branch (base already carries them). * fix(#2717): codex ESM-root behavioral test + hooks-built provenance for package.json Two review-driven follow-ups on the #2717 fix: - Adversarial review noted the ESM-root behavioral test covered only cursor; refactor it into a helper and add a codex case (the !isCodex-gated path most likely to regress, whose marker write lives in bin/install.js). gsd-check-update.js require()s at module load, so it surfaces the ESM failure immediately. - emitted-provenance flagged hooks/package.json as 'attributed source does not exist' — the marker is code-derived (a fixed literal emitted by ensureCommonJsMarker at install time), not built from a tracked source. Route the hooks-built rule's sources/transforms for package.json to the surface source file, mirroring the existing .cmd-shim sub-family. * chore(#2717): drop now-redundant hooks/package.json attribution ack The hooks-built provenance routing (prior commit) now self-attributes the emitted hooks/package.json to src/runtime-hooks-surface.cts, which IS in this diff — so the attribution is self-explaining and the emitted-drift-ack entry became stale. Delete the (now-empty) ack file per ADR-2719's empty-file rule. * docs(changeset): backfill #2717 PR number to 2846 |
||
|
Tom Boucher
|
b12d4df03b |
fix(#2694): normalize CRLF before frontmatter-boundary match in code-review workflows (#2839)
* test(#2694): CRLF frontmatter boundary regression for code-review workflows The code-review / code-review-fix workflows embed inline node -e one-liners whose frontmatter boundary regex used a literal \n, silently returning null on CRLF-saved SUMMARY.md / REVIEW.md / REVIEW-FIX.md artifacts and dropping every file in that summary (acceptance: per-artifact, no warning when the phase aggregate stays non-zero). Adds: - behavioral CRLF==LF boundary extraction tests (replica of the shipped one-liner's boundary step), proving the buggy literal-\n returns null on CRLF while the fixed normalize-then-match yields a byte-identical body; - a structural-regression-guard (allow-test-rule: structural-regression-guard) that reads the two shipped workflow files and asserts every boundary site normalizes \r\n -> \n before matching, so a revert of the fix is caught. * fix(#2694): normalize CRLF before frontmatter-boundary match in code-review workflows The code-review and code-review-fix workflows embed nine inline node -e one-liners that extract YAML frontmatter via a boundary regex content.match(/^---\n([\s\S]*?)\n---/) The literal \n defeated any CRLF-saved artifact (\r between --- and the line terminator), so SUMMARY.md / REVIEW.md / REVIEW-FIX.md saved with CRLF endings silently contributed zero files (or 'unknown' status / 'invalid') with no per-artifact warning. The Tier-3 git-diff fallback only fires when the aggregate across all summaries is zero, so a single CRLF summary among LF summaries produced no signal at all. Normalize \r\n -> \n once before the existing boundary match at all nine sites (code-review.md x3, code-review-fix.md x6). Byte-identical to the LF path; mirrors the canonical src/frontmatter.cts extractFrontmatter intent (CRLF == LF at the boundary); zero risk of \r leaking into field values consumed by the inner JS or the shell grep/cut pipeline. RED @ 94d0213 (3 failures, structural guard caught the shipped-text bug, both linux-node22+24 lanes).GREEN pending. * chore(#2694): acknowledge code-review workflow growth + changeset fragment emitted-attribution (ADR-2719) reports the byte growth from the CRLF-normalize insertion in code-review.md (+69) and code-review-fix.md (+138); both are the intended #2694 fix. Adds the .changeset Fixed fragment (pr:0, backfilled post-PR). * test(#2694): mixed CRLF/LF phase yields the union of both artifacts (criterion 2) The spec-axis review flagged that acceptance criterion 2 (a phase with a mix of CRLF-affected and unaffected artifacts no longer silently drops the CRLF artifact's contribution) was only transitively satisfied. Adds an explicit mixed-phase test replicating the full shipped Tier-2 extractor (boundary + inner key_files parse) across one LF and one CRLF SUMMARY.md, asserting the union of both — plus a RED proof showing the buggy boundary drops the CRLF artifact silently (aggregate non-zero, so the Tier-3 eq-zero fallback never fired). Locks the silent-partial-masking behavior the triage named as the more serious half of the defect. * docs(changeset): backfill #2694 PR number to 2839 * fix(#2694): make the CRLF regression test itself CRLF-lint-clean CI lint-tests caught that the new test tripped local/no-crlf-fragile-split: - the frontmatter boundary regex replicas (fixed + buggy) were RegExpLiterals with a bare \n; the rule flags frontmatter-shape regexes unconditionally. Build them via new RegExp(...) (byte-identical .source to the shipped literal) so the faithful replica is not a lint violation — the buggy replica MUST keep the literal \n, that is the bug it demonstrates. - the structural guard's src.split('\n') on the readFileSync'd workflow file was genuinely CRLF-fragile; use /\r?\n/ per the rule's canonical fix. - the allow-test-rule annotation gains its (#2694) tracking ref per ADR-456. lint:ci now exit 0 (incl. lint-allow-test-rule-refs, lint-emitted-drift-ack, lint-fix-has-regression-test: PASS). |
||
|
Tom Boucher
|
1fc21cdee0 |
fix(#2716): route non-user-facing conventional types to an Internal bucket, omit from release notes (#2838)
* test(#2716): failing-first regression for non-user-facing types → Internal bucket * fix(#2716): route non-user-facing conventional types to an Internal bucket, omit from release notes * fix(#2716): update SAMPLE_BODY/Discord/property tests for Internal bucket; fix stale CONTRIBUTING sentence (review) * test(#2716): relax Discord Enhancement assertion (enhance: prefix not stripped by cleanBullet) * docs(changeset): #2716 non-user-facing types omitted from release notes * docs(changeset): backfill #2716 PR number to 2838 |
||
|
Tom Boucher
|
6a9babda69 |
chore(#2798): declare the eleven reviewer lanes as manifest data (#2837)
* chore(#2798): declare the eleven reviewer lanes as manifest data Phase 5a of epic #2782, delivering ADR-2782 D9 (roster half) and D3. - Five reviewers GSD never installs into become lane-only role:reviewer capabilities with no runtime body, no runtimeCompat and no install surface: gemini, coderabbit, ollama, lm-studio, llama-cpp. Before this they had no descriptor at all and lived as a hardcoded NON_RUNTIME_REVIEWER_SLUGS tail, which is now deleted outright. - The six hosts that are ALSO reviewers gain a reviewer body alongside their runtime body. Their runtime bodies are byte-identical to next -- verified per capability against the git blob, not asserted -- so no install behaviour moves. - KNOWN_REVIEWER_SLUGS derives from declared bodies via an exported deriveReviewerSlugs(registry). hostBehaviors.reviewerCli survives as a derived legacy alias for one release; where a capability carries both, the body wins and the slug appears once. Alias removal is Phase 7 (#2801). THE KEYSTONE: the roster is the SAME ELEVEN SLUGS as before -- antigravity, claude, coderabbit, codex, cursor, gemini, llama_cpp, lm_studio, ollama, opencode, qwen. This phase changes HOW the roster is derived, not WHO is in it, and the test asserts that literal list rather than a count. kimi-code is deliberately NOT declared here. It is net-new with no invoke_reviewers leg, so declaring it now would make it selectable but not invocable -- present in --all, selected, emitting an empty section for the whole 5a-to-5b window -- and would break Phase 1's parity assertion. It lands in 5b alongside the iteration that can run it. Legacy kimi (the Python CLI) is not a reviewer at all and gains nothing. The highest-value test is declaredManifestLanesMatchThePhase1Descriptor: it deep-compares all eleven declared bodies against REVIEWER_LANES field-by-field, including probe and invoke sub-fields. All eleven are byte-identical, key order included. The epic's premise is that the manifest and the core descriptor describe the same lane with NO translation layer, and Phase 2's review already caught one divergence that every other test missed. Two ADR corrections folded in, as Phases 1-3 each did: 1. PHASE ORDER. The ADR runs Phase 4 (federated config) before 5a and #2798 claims a dependency on 4. That is inverted and makes Phase 4 unsatisfiable: D9 assigns review.<host>_host to lane capabilities that do not exist until THIS phase creates them, and a federated config slice must live inside capabilities/<id>/capability.json. Real graph: Phase 2 -> 5a -> 4. 2. #2798's INVENTORY acceptance item is vacuous. The inventory catalogs bin/lib/*.cjs modules, not capability directories -- antigravity, opencode and qwen appear zero times in it -- and gen-inventory-manifest --check passes with the five new dirs and no edit. Also corrected a stale line in Phase 2's own ADR amendment: it recorded the slug pattern as /^[a-z][a-z0-9_-]*$/, but Phase 2's security review widened the shipped pattern to /^[a-z0-9][a-z0-9_-]*$/ to match Phase 1's exported LANE_SLUG_RE. The prose had not followed the code. Closes #2798 * fix(#2798): catalogue reviewer capabilities in the generated matrix The capability matrix rendered exactly two tables, feature and runtime, via renderTable(caps, role) filtering on c.role === role. ADR-2782 D3 added a THIRD role, so every role:"reviewer" capability was silently dropped from the first-party catalogue. The drift guard did not catch it, and could not: --check compares generated output against the committed file, and both omitted the five lanes identically, so it reported "up to date" while five shipped capabilities were invisible in the one document that is supposed to list what ships. A guard blind to an entire role is not guarding. This phase is what exposed it -- it ships the first role:"reviewer" capabilities -- so it is fixed here rather than deferred (CLAUDE.md: a defect found while working is fixed in the current change, which overrides one-concern-per-PR). Verified red-before-green: with a lane row deleted from the matrix, --check now exits 1; restored, it exits 0. Before this fix the lanes were absent entirely, so there was nothing for the guard to compare. Phase 6 (#2800) still owns enriching the matrix with lane-specific detail (slug/flag/transport columns) and the locale parity gate. This is the narrower fix: the capabilities APPEAR at all. * fix(#2798): close two hardening gaps and record three limits durably Isolated security review (5 targets, no blockers) reproduced two gaps in the new deriveReviewerSlugs. Both are unreachable through the checked-in registry -- it is generated, JSON-sourced and code-reviewed -- but the function is EXPORTED for reuse and carries no other validation, so it must not depend on its caller. - A whitespace-only slug passed the length>0 test verbatim and occupied a roster entry it could never match. Slugs are now trimmed before the emptiness test. A blank body correctly falls through to the legacy alias rather than DROPPING the lane, which would have been worse than the blank slug. - KNOWN_REVIEWER_SLUGS is computed at require() time, so an uncaught throw there breaks import for EVERY consumer rather than degrading selection. It is now guarded, yielding an empty roster on a malformed registry. That is a visible degradation, not a silent one: under D4 an explicitly requested reviewer that is unavailable is an ERROR, so /gsd:review --claude against an empty roster fails loudly. This also removes an asymmetry -- the sibling capability-trust module documents its collectors as TOTAL and wraps them for exactly this reason. Also records three findings that previously existed ONLY in squash-merged PR bodies, which is not a durable record: - ADR-2782 D5 gains an implementation note explaining why the resolved host is deliberately EXCLUDED from the disclosure signature. Rule 1 says consent binds the resolved host; the loader has no config resolver, so folding it in would make the loader and lifecycle compute different signatures for one manifest and re-prompt forever. The binding is split: signature covers the SHA-pinned manifest fields, the consent record stores the resolved host, and Phase 5b re-resolves at invocation -- which is where rule 4 already puts the check. A reader comparing rule 1 to the code would otherwise conclude it is unimplemented. - CONTEXT.md's capability-trust entry still described THREE executable surfaces. Phase 3 added the fourth and made that false; corrected here, since it is drift this epic introduced rather than Phase 6's new-glossary-term work. - stableJson documents the NaN/Infinity/undefined -> null signature collision and why it is unreachable (JSON grammar has no such literal, so JSON.parse throws first). Reachability rests entirely on the ingest path staying JSON.parse-only, so the note lives where someone would break it. * chore(#2798): backfill changeset pr number to 2837 |