mirror of https://github.com/garrytan/gstack.git
258 Commits
| Author | SHA1 | Message | Date |
|---|---|---|---|
David Grant
|
b39da77129 |
test: simplify matchesUnnegated, drop its dedicated unit test file
Per PR review feedback: the negation-aware matcher these E2E evals rely on had accumulated real complexity (sentence-boundary scanning, markdown-bullet clause detection, curly-apostrophe contraction matching, idiom exclusions) across four rounds of specialist/adversarial review during this branch's own /ship — each finding was real, but chasing all of them was disproportionate for a periodic-tier, non-gating eval. An occasional missed edge case there just means a rare, visible false read, not a missed real regression. Reverted to a plain fixed-window lookback (60 chars, accounting for the patterns' own internal gaps) with a short negation word list. Kept the generic [a-z]+n't contraction match (can't/won't/aren't) since it's one regex alternative, not meaningful added complexity. Removed test/helpers/e2e-helpers.test.ts — a dedicated unit-test file for a best-effort text-matching heuristic used only by 4 paid, periodic-tier evals was more test infrastructure than the actual ask (3-4 new eval cases) warranted. |
|
|
David Grant
|
54a50e4b67 |
fix(plan-eng-review,plan-ceo-review): trim scope per PR review — drop CEO bullets, own section, fewer patterns
PR review feedback on #1071: - Dropped the data-model exception + JSONField bullets from plan-ceo-review entirely. CEO review should stay high-level on "right-sized diff"; data-model pushback belongs to plan-eng-review, which already carries the full checklist. plan-ceo-review/SKILL.md(.tmpl) are now byte-identical to origin/main. - Trimmed both remaining plan-eng-review bullets (data-model exception, JSONField) to single-sentence terseness, matching the style of sibling preferences instead of reading like a paragraph. - Removed cognitive patterns 11-13 (normalize-first, SRP-for-models, structure-beats-blobs) from the numbered "Cognitive Patterns" list — they were more tactical/lower-level than the rest of that list (Larson, McKinley, Conway-style instincts), and duplicated what the Data Model Review now covers operationally. Renumbered 14-18 down to 11-15. - Promoted the Data Model Review Checklist from a "####" subsection tacked onto Architecture Review to its own "### Data model review" section with its own AskUserQuestion/STOP gate, matching Architecture/Code Quality/Test/Performance as a peer rather than an afterthought. Did not renumber the existing 1-4 sections (Architecture/Code Quality/Test/ Performance) since those numbers are referenced elsewhere (carve-guards.ts, the main SKILL.md.tmpl's section-order prose). Updated test/skill-validation.test.ts's static guardrails to match: removed the plan-ceo-review assertions (nothing left to guard there), removed the three now-deleted cognitive-pattern assertions, updated the contiguity check from 1-18 to 1-15, and updated the checklist-subsection test to check for the new "### Data model review" heading. Tightened test/helpers/carve-guards.ts budgets back down now that content shrank: plan-ceo-review reverted fully to its original 90_000/1.08 (content is byte-identical to baseline again); plan-eng-review tightened from 70_000/1.15 to 67_500/1.13 (down from the prior over-generous bump, though not fully back to the original 67_000/no-ratio since the trimmed bullets are still a small net addition). bun test test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts test/gen-skill-docs.test.ts — all pass. |
|
|
David Grant
|
1820f58f69 |
fix: drop bare "no" as a negation trigger, treat markdown bullets as boundaries
Codex structured review (large-diff gate, Step 11) found 3 more real gaps:
- matchesUnnegated's bare "no" trigger over-fired on common positive-framing
idioms: "No blocker here: create a separate tier model" and "there is no
downside to promoting the payload" both got misread as negated, even
though they're endorsements of the recommendation that follows, not
rejections. Removed bare "no" from the negation list — "not"/"never"/
"cannot"/the "-n't" family are far more reliable negation signals and
don't share this idiom-collision problem.
- SENTENCE_END only recognized period-followed-by-capital as a boundary, so
a bulleted recommendation like "I would not keep this inline.\n- Separate
tier model" (very common LLM output formatting) let the earlier "not"
leak into the bullet below it. Added a second alternative: a newline
followed by a markdown bullet or numbered-list marker now also counts as
a clause boundary. Fixed the boundary-index arithmetic to use the full
match length (em[0].length) instead of a bare +1, since this new
alternative can match more than one character ("\n- ", "\n1. ").
- The minimal-change eval's own hardcoded doesn't-pattern only matched the
ASCII apostrophe (`doesn'?t`), missing the curly one — inconsistent with
matchesUnnegated's own apostrophe handling a few lines away in the same
file.
Added 3 regression tests reproducing each directly.
bun test test/helpers/e2e-helpers.test.ts test/skill-e2e-plan.test.ts
test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts
test/gen-skill-docs.test.ts — 788 pass, 42 skip (paid E2E), 0 fail.
|
|
|
David Grant
|
02e312c057 |
fix: harden data-model E2E fixtures against reading the operator's real gstack install
Codex adversarial review (Step 11) flagged that these 4 tests were not
hermetic: the prompt only said "Read plan-eng-review/SKILL.md" (relative,
unpinned), and the copied SKILL.md's own STOP-Read directive references an
ABSOLUTE path (~/.claude/skills/gstack/plan-eng-review/sections/review-sections.md)
for its carved sections file. On any machine with gstack actually installed
at that location — which includes the maintainer's own dev machine — the
agent could read and get graded against the REAL globally-installed skill
instead of this branch's sandboxed copy, silently defeating the point of
the isolated fixture. This is the exact failure mode
test/helpers/auq-sdk-capture.ts's setupSkillDir()/captureSectionReads()
already guards against for other carved-skill tests; these 4 blocks
predated that pattern and didn't use it.
Fixed with a targeted change rather than a full migration to the
auq-sdk-capture harness: extracted planEngReviewDataModelPrompt() into
e2e-helpers.ts, pinning the sandboxed SKILL.md's absolute path and
explicitly forbidding the agent from reading any other SKILL.md ("especially
nothing under ~/.claude or /Users") — matching auq-sdk-capture.ts's own
phrasing. This also resolves the maintainability specialist's earlier DRY
finding (the identical prompt string was duplicated across all 4 blocks) in
the same change, since all 4 now call the one shared function.
Also fixed a second Codex finding: touchfiles.ts only listed
plan-eng-review/** as these tests' dependency, but their actual behavior
lives partly in test/helpers/e2e-helpers.ts (setupPlanEngReviewFixture,
matchesUnnegated) — a future bug there wouldn't have gotten these 4 tests
re-selected by the diff-based selector. Added the file to each test's
touchfile entry (scoped to just these 4, not GLOBAL_TOUCHFILES, since the
new exports aren't used elsewhere yet).
Two other Codex findings reviewed and accepted as-is, not fixed:
- matchesUnnegated can still mis-score idiom-based false negation ("no
doubt", "not only... but also") — a genuine lexical-heuristic limitation,
but these idioms are exceedingly unlikely to co-occur with the specific
verb+object patterns these checks require in real review prose; chasing
every conceivable idiom is diminishing returns for a best-effort matcher.
- The new evals are periodic tier, not gate — consistent with this
repo's existing convention for all Opus-model E2E tests (documented in
CLAUDE.md's two-tier system), not a gap introduced by this branch.
bun test test/helpers/e2e-helpers.test.ts test/skill-e2e-plan.test.ts
test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts
test/gen-skill-docs.test.ts — 786 pass, 42 skip (paid E2E), 0 fail.
|
|
|
David Grant
|
b3acffdf58 |
fix: catch can't/won't-style contractions, guard data-model-bias against negation
Maintainability specialist re-dispatch found the standalone n't alternative
in NEGATION was unreachable: \b n't \b can never match mid-word (no word
boundary exists between the letters immediately before "n" and "n" itself
in "can't", "won't", "aren't", "hasn't"), so only the explicitly-spelled-out
contractions (don't, doesn't, didn't, shouldn't, wouldn't, isn't) were ever
detected — very common ones like "can't"/"won't"/"cannot" fell through
silently. Replaced with a generic [a-z]+n't pattern plus an explicit
"cannot" alternative (which has no apostrophe to match generically).
Also brought the original data-model-bias test's recommendsSeparateModel
check in line with its three newer sibling counterexample tests by routing
it through matchesUnnegated — it was the one E2E block still un-guarded
against a negated false-positive ("I would NOT recommend a separate tier
model" would previously have registered as a pass).
Did not touch rejectsAsUnjustified (measured-denorm test) — it directly
models rejection language ("instead of denormalizing" IS the signal, not a
positive claim needing negation-checking), so wrapping it in
matchesUnnegated would conflict with its own tuned alternatives for no
clear benefit. Did not extract a shared prompt-string constant across the
4 E2E blocks (maintainability finding) — the identical prompt string already
appears verbatim in 3 pre-existing, untouched blocks in this same file;
extracting a constant for only the 4 new blocks would create inconsistency
rather than resolve it, and touching the pre-existing blocks is outside
this PR's scope.
Added unit tests reproducing both contraction gaps directly.
bun test test/helpers/e2e-helpers.test.ts test/skill-e2e-plan.test.ts
test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts
test/gen-skill-docs.test.ts — 783 pass, 42 skip (paid E2E), 0 fail.
|
|
|
David Grant
|
aa49339b27 |
fix: widen matchesUnnegated to sentence boundary, catch contrastive phrasing
Ship-workflow review re-dispatched the testing specialist against the full
updated diff (including the new minimal-change eval) and found 3 more real
bugs in the same family as the earlier fix pass:
- matchesUnnegated's fixed 20-char negation lookback was too narrow: the
patterns it guards have their own internal gaps up to 80 chars, so "I do
not think it's worth extracting the payload into columns" (negation >20
chars before the match) would be misread as an unnegated recommendation.
Now scans back to the start of the current sentence (last ./!/? before the
match) instead of a fixed count — matches the same period-bounded
assumption the calling patterns already make.
- Verified during the fix: neither "not"/"never"/etc. nor the wider window
catches contrastive phrasing ("add this field to the model RATHER THAN
create a separate table") — the rejected alternative matches the pattern
just as strongly as a genuine recommendation, with no negation word
anywhere nearby. Added "rather than"/"instead of" to the shared negation
list, benefiting all four callers, not just the new eval.
- The new minimal-change eval's acceptsInlineAddition check was missing the
matchesUnnegated guard entirely on its first alternative (unlike its
sibling recommendsSeparateModel a few lines above), and
recommendsSeparateModel's verb list missed common recommendation phrasings
(introduce, build, spin out, break out) — deliberately did NOT add "add",
since "add this new field to the existing model" is exactly how the
CORRECT answer gets phrased.
Added 3 unit tests reproducing the sentence-boundary and contrastive-phrasing
fixes directly (not just via the paid E2E path).
bun test test/helpers/e2e-helpers.test.ts test/skill-e2e-plan.test.ts
test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts
test/gen-skill-docs.test.ts — 781 pass, 42 skip (paid E2E), 0 fail.
|
|
|
David Grant
|
a9124bd15b |
test: add minimal-change counterexample eval, unit-test matchesUnnegated
garrytan/gstack#1048 comment asked for concrete evaluations covering three
cases: normalized models, justified JSON fields, and minimal-change cases.
The prior commits covered the first two (data-model-bias, legitimate-json,
measured-denorm) but not the third — a plan where keeping a single trivial
field inline (not extracting a new model) is the correct call. This was
independently flagged by the ship-workflow coverage audit as the highest-
severity gap in this branch's test coverage.
Adds plan-eng-review-data-model-minimal-change: a synthetic plan adding one
nullable timestamp field to an existing model, no polymorphism, no JSON, no
independent query pattern or consumer. Asserts the skill does NOT reflexively
recommend extracting a separate model for it — exercising the "unless the
extra job is a single trivial field..." exception added to cognitive pattern
#12 and the SRP checklist item in
|
|
|
David Grant
|
09ad6fb1ef |
fix: harden new data-model eval regexes, dedupe E2E fixture setup
Ran /review on this branch's diff against main. The always-on testing
specialist found real correctness bugs in the two eval blocks added in
33fbb58b:
- pushesToSplitTheField (legitimate-JSON eval) was negation-blind: a
CORRECT "I would NOT extract the payload into columns" answer contains
the same verb+payload+column proximity as an incorrect one, so it could
fail the test on a passing review. Also missing "normalize" from its verb
list, so a real regression phrased as "normalize this payload into
columns" would slip through undetected.
- rejectsAsUnjustified (measured-denorm eval) missed common rejection
phrasings ("instead of denormalizing", "rather than snapshotting"), and
its normali[sz]e-instead/back/it clause was loose enough to false-positive
on approving language ("stays normalized elsewhere; keep it that way").
- recommendsSeparateModel (pre-existing data-model-bias eval, touched by
this branch to fix its fixture) broke on backtick-wrapped model names, a
common LLM markdown convention.
Added matchesUnnegated() to test/helpers/e2e-helpers.ts — walks every regex
match and checks the ~20 chars before it for a negation word before counting
it as a positive signal — and used it for the split-detector. Broadened the
rejection regex and tightened its false-positive-prone clause. Stripped
markdown formatting before the backtick-affected check. Verified all four
fixes against representative sentences (correct-negative, actual-positive,
previously-missed-regression, and previously-false-positive cases) before
committing — all passed as expected.
Also extracted the maintainability specialist's DRY finding: the three
data-model E2E blocks (bias, legitimate-json, measured-denorm) each
duplicated the same ~20-line mkdtemp/git-init/write-plan/copy-skill fixture
setup. Extracted into setupPlanEngReviewFixture() in e2e-helpers.ts.
bun test test/skill-validation.test.ts test/parity-suite.test.ts
test/touchfiles.test.ts test/gen-skill-docs.test.ts test/skill-e2e-plan.test.ts
— 768 pass, 39 skip (paid E2E), 0 fail.
|
|
|
David Grant
|
33fbb58b13 |
test: resolve terminology drift, rebase E2E fixture to carved templates, add positive/counterexample evals
PR review feedback (garrytan/gstack#1071), resolving the three Copilot review threads plus the "add positive and counterexample evaluations" ask: - test/skill-validation.test.ts: renamed the two tests still named "...to minimal diff..." to "...to right-sized diff..." to match the actual bullet text the assertions check (Copilot review comment). - test/skill-e2e-plan.test.ts: updated the Data-Model Bias regression test's header comment from "minimal diff" to "right-sized diff" (Copilot review comment); fixed the same test's fixture setup to also copy plan-eng-review's sections/ directory — main's carve refactor moved the Architecture Review and Data Model Review Checklist out of SKILL.md into sections/review-sections.md after this test was written, so the sandbox was missing the very file the skill's STOP-Read directive points at. - Added two new periodic-tier E2E cases exercising the counterexamples the templates now document: "plan-eng-review-data-model-legitimate-json" (a Stripe-webhook-style JSONField the skill should NOT push back on) and "plan-eng-review-data-model-measured-denorm" (a scoped denormalization backed by a stated APM/load-test measurement the skill should NOT insist on normalizing away). Both registered in test/helpers/touchfiles.ts (periodic tier, same classification as the existing bias regression test). Verified: bun test test/skill-validation.test.ts test/parity-suite.test.ts test/touchfiles.test.ts test/skill-e2e-plan.test.ts — 768 pass, 39 skip (paid E2E, requires EVALS=1), 0 fail. |
|
|
David Grant
|
5374987c6c |
fix(plan-eng-review,plan-ceo-review): narrow absolute schema-bias language, add legitimate-JSON and measured-denormalization counterexamples
PR review feedback (garrytan/gstack#1071): the data-model bias guidance was
overly absolute ("diff size is NOT a goal", "three clean models... in all
respects", "JSONField is not an escape hatch") and lacked counterexamples for
cases where JSON columns or denormalization are the right call.
- "Data model exception to right-sized diff" bullet (both skills): reframes
from an absolute rule to "don't let diff size alone decide the shape",
explicitly allows denormalization backed by a measurement, and adds the
counterexample of a field that would only ever be read through one FK join
with no independent consumer.
- JSONField bullet (both skills): keeps the core diagnostic but adds concrete
legitimate uses — third-party webhook/OAuth payload caching, open-ended
preference bags, event payloads shaped by an external producer, and
schemas still being discovered.
- plan-eng-review cognitive patterns #11 (normalize first) and #13
(structure beats blobs) get the same measured-reason / known-polymorphism
counterexamples.
- Architecture Review's "Data model honesty" bullet and the "Single
Responsibility" and "JSONField as polymorphism escape hatch" checklist
items (now living in sections/review-sections.md.tmpl per main's carve
refactor) get matching nuance.
Bumps carve-guards.ts skeleton-size budgets for plan-eng-review (69_000 ->
70_000) and plan-ceo-review (91_000 -> 92_000) to cover the added prose;
verified against test/parity-suite.test.ts (13/13 pass) and
test/skill-validation.test.ts's load-bearing-bullet guardrails (still pass —
none of the anchor phrases the static tests check were removed).
|
|
|
David Grant
|
fe3f0d6d6c |
merge: sync upstream main (v1.60.2.0)
Merges origin/main ( |
|
t
|
4ad55e2eef | Merge remote-tracking branch 'origin/main' into time-attack/wave-test-infra-isolated-review | |
Joshua France
|
2beb636f7c
|
Merge pull request #2234 from time-attack/time-attack/issue-2194-pglite-lock
fix(gbrain): distinguish PGLite locks from malformed config |
|
|
Joshua France
|
050abc7aae
|
Merge pull request #2222 from time-attack/fix-brain-trust-policy-reject
fix(gstack-config): accept @local endpoint ids for brain_trust_policy |
|
|
Joshua France
|
15c069631b
|
Merge pull request #2221 from time-attack/fix-gstack-issue
fix(benchmark): gemini stream-json content/stats parse (#2159) |
|
|
Joshua France
|
e284c8e0cd
|
Merge pull request #2218 from time-attack/plan-daily-gstack-pr
fix(repo-mode): reuse canonical cached project slug (#2212) |
|
|
t
|
145fe2d4d1 | fix(ios-qa): make SwiftUI device taps observable and reliable | |
|
t
|
8e25d1583d | fix(ios-qa): generate app-owned bridge accessors deterministically | |
|
t
|
ea648b7dff | fix(careful): fail closed for compound recursive deletes | |
|
t
|
a2a447a117 | fix(safety): classify credential actions consistently | |
|
t
|
239d73afc4 | fix(hosts): validate suppressed resolver names | |
|
t
|
0b869898ed | fix(ios-qa): make generated Swift manifests valid on first run | |
|
t
|
1aa5283ddd | fix(gbrain): canonicalize remotes with trailing slashes | |
|
t
|
7e23aecfc5 | fix(gbrain): parse nested context query filters | |
|
t
|
45d0fb2b54 | fix(eval): validate eval list limit | |
|
t
|
aa3bd6f042 | fix(test-harness): spawn claude without shell interpolation | |
|
t
|
5cb8761d36 | fix(test-harness): measure first response on first NDJSON line | |
Sinabina
|
cbf8a2bec2 | fix(gbrain): classify PGLite connect locks | |
|
t
|
8b53302983 |
test(deps): tripwire that no basic-ftp < 5.3.1 resolves in bun.lock
Parses package.json overrides + every resolved basic-ftp specifier in bun.lock (including nested paths like get-uri/basic-ftp) and fails if any is below 5.3.1. Deterministic and offline. Fires on the pre-fix tree (basic-ftp@5.2.0) and would also catch a direct-dependency-only bump that leaves a nested vulnerable copy. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
Sina
|
db149a6162 |
fix(gstack-config): accept @local endpoint ids for brain_trust_policy
endpoint-hash returns "local" for stdio/PGLite, but validators only allowed hex suffixes, so setup-gbrain could not persist trust policy. Co-authored-by: Cursor <cursoragent@cursor.com> |
|
|
Sina
|
c7e182baa2 |
fix(benchmark): skip-trust for headless gemini + drop release metadata
Pass --skip-trust so temp/untrusted workdirs work in headless benchmarks (mirrors Codex --skip-git-repo-check). Revert VERSION, CHANGELOG, and package.json bumps — leave release bookkeeping to maintainers / /ship, matching community PR practice. Co-authored-by: Cursor <cursoragent@cursor.com> |
|
|
Sina
|
5d93ce278d |
test(benchmark): full post-CLI gemini e2e + GEMINI_API_KEY auth path
Exercise resultFromGeminiStream against current stream-json fixtures (including empty-success hardening). Recognize GEMINI_API_KEY and map IneligibleTierError to auth — personal OAuth free-tier is no longer supported by gemini CLI. Co-authored-by: Cursor <cursoragent@cursor.com> |
|
|
Sina
|
58e9881d6e |
fix(benchmark): parse current gemini stream-json content/stats shape
GeminiAdapter was reading message.text and result.usage, so current CLI content/stats events produced empty $0 success rows. Accept content with an assistant role guard, stats token fallbacks, init model, and treat empty exit-0 output as an error (#2159). Co-authored-by: Cursor <cursoragent@cursor.com> |
|
|
Sina
|
8aab6dfb7e |
fix(repo-mode): reuse canonical cached project slug (#2212)
gstack-repo-mode was re-deriving slugs from origin and writing twin project dirs; consume gstack-slug's cached sanitized slug instead. |
|
Garry Tan
|
88dec06907
|
fix(test): autoplan-dual-voice sandbox registers skills + assertions read real stream-json
Root cause of months of silent local failure: the sandbox copied skill dirs to the repo root, but claude >= 2.x resolves slash commands strictly from registered skills, so /autoplan short-circuited with 'Unknown command' (0 turns, ~1s) on every attempt. Install /autoplan + review skills at project-level .claude/skills/ (same pattern as skill-routing-e2e). Also: the transcript filter matched entry.type === 'tool_use', a shape that never appears at the top level of raw stream-json, so assertions only ever saw the final result text; filter on assistant/user events instead. Hang protection accepts the Phase 1 review dispatch (Agent/Task tool call carrying review instructions) as progress evidence, since full Phase 1 completion is 15+ min of subagent work. Budget raised to 10 min / 40 turns. Invisible in CI: the file is in neither evals.yml nor evals-periodic.yml matrices (coverage decision filed in TODOS.md). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
|
|
Garry Tan
|
bb4de3926f
|
fix(test-harness): return promptly when a timed-out child leaves pipe-holding orphans
proc.kill() only signals the sh -c wrapper; the claude child survives as an orphan that inherited our stdout/stderr pipes, blocking the stream drain until it exits (observed: a 600s spawn timeout stretching to 1431s and tripping bun's per-test timeout with no result). On timeout, cancel the stdout reader; race the stderr drain against child exit + 5s grace. Streamed transcript lines survive the cancel, so callers still get their evidence. Regression test: test/session-runner-timeout.test.ts (fake claude spawns a pipe-holding orphan; fails in 30s without the fix, passes in 8s with it). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> |
|
|
Garry Tan
|
11de390be1
|
v1.58.5.0 feat: first-run activation scaffold + gstack router front door (#2078)
* feat: first-run activation — project-aware scaffold, router front door, onboarding nudges Adds the activation system that drives a new install toward a concrete first move: - bin/gstack-first-task-detect: local-git+filesystem repo classifier emitting one validated enum bucket (greenfield/code_<lang>/branch_ahead/dirty_default/clean_default), portable timeouts, fail-safe empty output. - generate-first-run-guidance.ts: unified preamble section — first-run project-aware scaffold + returning-session plan->review->ship tip, gated on a persistent .activated marker and never run in headless. Detection wired lazily in generate-preamble-bash.ts. - SKILL.md.tmpl: top-level gstack skill is now a pure router (browse body removed; it lives in /browse), routing any request and sending browser/QA work to /browse. - setup: first-move nudge on first install. office-hours: closing handoff that launches the next review via the Skill tool. - telemetry-ingest: accept onboarding/first_task_scaffold_shown/handoff/route event types. * test: cover first-run detection + repoint browse-content assertions to /browse - New unit tests for every detection bucket, the eval-safe enum contract, and the first-run gating (test/preamble-first-task-scaffold.test.ts); periodic E2E that runs the detector through the real harness (test/skill-e2e-first-task-scaffold.test.ts). - Repoint browse-content assertions (gen-skill-docs, audit-compliance, skill-validation, LLM-judge eval) from the root skill to browse/SKILL.md following the router split; add a regression pinning that the router carries no browse body. - Register first-task-scaffold touchfiles + periodic tier; bump parity/carve size caps ~1-2KB per skill for the shared first-run-guidance preamble section. - Refresh ship golden fixtures for the preamble addition. * chore: regenerate SKILL.md + llms.txt for first-run activation * chore: bump version and changelog (v1.58.5.0) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(test): repoint bws skillmd-* setup-block assertions to browse/SKILL.md The skillmd-setup-discovery / -no-local-binary / -outside-git E2E tests extracted the `## SETUP`→`## IMPORTANT` browse binary-discovery block from the root SKILL.md. P2 moved that block to browse/SKILL.md (end anchor is now `## Core QA Patterns`), so the slice came back empty and the `browse/dist/browse` guard failed. Repoint to browse/SKILL.md. Verified: 7/7 e2e-browse pass locally. * fix(test): tolerate skill-discovery race in PTY plan-mode smoke The e2e-pty-plan-smoke suite (office-hours / plan-mode-no-op) failed in CI with `Unknown command: /office-hours` (claude exited ~10s) while passing locally. Root cause: a cold CI container's overlay-FS scan of the symlinked ~/.claude/skills registry finishes AFTER the runner's 8s boot grace, so the first `/skill` send reaches claude before the skill is indexed and is rejected as unknown. The runner gave up on the first "Unknown command:" line. runPlanSkillObservation now re-sends the skill command up to 3x (6s apart), re-marking the buffer each time so stale scrollback can't re-trip the check, before concluding the skill is genuinely unregistered. A real dangling-symlink / missing-skill still surfaces as 'exited' (after retries), preserving the original diagnostic. Pure-helper contract unchanged: 95/95 unit tests pass. This is a pre-existing harness bug (fails identically on #2077's own branch, which introduced the suite) surfaced while shipping the activation feature. * debug(ci): temporarily instrument pty-smoke skill discovery Capture claude version, env, registry tree, and a claude -p discovery probe to pin why /office-hours isn't discovered in CI (retries proved it's not a race). Temporary — revert once the registry fix is identified. * chore: revert pty-smoke harness experiments (race-retry + CI debug step) Diagnosis is conclusive and the experiments aren't the fix, so restore the harness to its original state (net-zero diff vs main for both files). What the CI debug step proved: `claude -p` returns READY — claude v2.1.187 fully DISCOVERS /office-hours from the symlinked registry. Only the interactive PTY TUI rejects it as "Unknown command" (and it received the full command text). So the e2e-pty-plan-smoke failure is a claude 2.1.187 interactive-TUI regression (skills discovered by `claude -p` aren't exposed as TUI slash commands), pre-existing in the #2077 harness and failing identically on its own origin branch — unrelated to this activation PR. The race-retry can't help (the TUI genuinely lacks the command); the debug step also tripped actionlint (shellcheck SC2012). Both reverted. * fix(ci): copy SKILL.md as real files in pty-smoke registry (cross-mount symlink) The e2e-pty-plan-smoke suite failed with "Unknown command: /office-hours" in CI while passing locally. Root cause (proven, not guessed): claude 2.1.187's interactive-TUI skill scanner does not follow the /github/home -> /__w cross-mount symlink the registry used for per-skill SKILL.md. Evidence: a CI debug step showed `claude -p` discovered the skill (printed READY), and a local macOS repro with the identical symlinked registry recognized /office-hours — isolating the failure to the container's cross-mount symlink, not registration content, claude version, duplicate names, or a race. Fix: register the per-skill SKILL.md + sections as REAL copies (same mount as $HOME) so the TUI reads them directly. The gstack root stays a symlink — the preamble's runtime bash resolves bin/* and sections/* through it and bash follows cross-mount symlinks fine. * fix(ci): guard rm expansion in pty-smoke registry (shellcheck SC2115) * fix(ci): also register pty-smoke skills project-scoped (cwd/.claude/skills) The real-file user-dir registration still left the TUI rejecting /office-hours in the container. claude's interactive TUI surfaces /slash commands from the PROJECT dir (<cwd>/.claude/skills); the smokes run with cwd=$REPO whose .claude/skills is gitignored (absent on a fresh CI checkout), so the user-dir registry feeds `claude -p` (READY) but not the TUI. Populate $REPO/.claude/skills with real SKILL.md + sections copies (no gstack symlink there — it would point at its own parent; runtime paths use the user-dir gstack symlink). --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
9fd03fae9e
|
v1.58.4.0 fix: high-priority community bug wave + PTY plan-mode smoke gate (#2077)
* fix(gbrain): stop forcing GBRAIN_PREPARE on transaction-mode poolers (#1965) buildGbrainEnv auto-set GBRAIN_PREPARE=true whenever DATABASE_URL targeted port 6543, and the /sync-gbrain capability check exported it for the rest of the skill run. Both had the semantics inverted: gbrain auto-disables prepared statements on transaction-mode poolers because they break every write there ("prepared statement does not exist"); GBRAIN_PREPARE=true is gbrain's documented override for SESSION-mode poolers on 6543, not a requirement for transaction mode. The #1435 search symptom the auto-set worked around was fixed gbrain-side. Remove both force-sets. A caller-set GBRAIN_PREPARE (either value) still passes through untouched, preserving the session-mode-on-6543 escape hatch. isTransactionModePooler stays exported. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(gbrain): classify probe timeout as its own status; sync proceeds instead of skipping (#1964) The 5s engine probe misclassified healthy-but-slow engines (cold Supabase pooler connections measured at 6.9-10.7s) as broken-config, so /sync-gbrain silently skipped code+memory and told the user their config was malformed. - New "timeout" status: probe killed at the deadline with no recognized stderr pattern. Default deadline is now 15s, overridable via GSTACK_GBRAIN_PROBE_TIMEOUT_MS (tests set 300ms against a fake that sleeps 2s). - Sync stages PROCEED on timeout with a stderr warning naming the env knob; a genuinely-dead engine surfaces its real error at the first operation instead of a false config diagnosis. - Consistency everywhere "ok" gated behavior: gstack-gbrain-detect --is-ok exits 0 on timeout, and gen-skill-docs' detection gate accepts it, so a slow engine no longer silently suppresses brain-aware features. - Status cache: key now includes the effective probe timeout (raising it invalidates a cached timeout) and GBRAIN_HOME; config detection honors GBRAIN_HOME so relocated-home users stop being misclassified as missing-config. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(bins): cygpath-normalize SCRIPT_DIR for bun imports; surface learnings-log errors (#1950) Under Windows git-bash, pwd yields a POSIX path (/c/Users/...) that Bun on Windows cannot resolve as an ES module specifier. gstack-learnings-log interpolates SCRIPT_DIR into a bun -e import, so every invocation died with "Cannot find module" — and 2>/dev/null swallowed the error, silently dropping every AI-logged learning for Windows users. - 3-line cygpath -m guard in gstack-learnings-log and gstack-question-log (which gains the same import shape in the next commit). Matches the duplicated IS_WINDOWS convention in setup; no shared shell lib exists. - learnings-log adopts question-log's set +e / TMPERR capture pattern wholesale: validation errors now print to stderr. The old `if [ $? -ne 0 ]` check was dead code under set -euo pipefail — the script exited at the failing assignment before reaching it. - New test/bin-windows-bun-import-paths.test.ts: static invariant (any bash bin interpolating $SCRIPT_DIR into a bun -e import must carry the guard) + behavioral end-to-end run invoked via `bash <bin>` — added to the windows-free-tests workflow list so the conversion is proven on the only platform where the bug exists. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(question-log): dedupe INJECTION_PATTERNS via lib/jsonl-store (#1934) bin/gstack-question-log carried a local copy of the injection-pattern list, so pattern fixes to lib/jsonl-store.ts never propagated — including the /override[:\s]/i false-positive fix arriving via community PR #1940. Import the shared hasInjection instead (enabled by the previous commit's cygpath guard). question-log also gets the lib's stricter superset (human:, disregard, from-now-on, approve-all patterns). Tests pin the contract in a #1940-order-independent way: an "Override: ignore all previous instructions" header is rejected, "prose overrides the deterministic table" is accepted, and a static invariant keeps local INJECTION_PATTERNS duplicates out of the bin. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(security): community-pulse + both dashboards never report fake zeros (#1947) The security-signaling surface failed open at three layers — every failure mode read as a reassuring "0 attacks" / "0 installs": - community-pulse edge function: supabase-js returns {data,error} without throwing, and all five queries discarded `error` — a DB outage produced real-looking zeros via the SUCCESS path, and the catch (also returning zeros with HTTP 200) was unreachable for query failures. Every query now destructures and throws; the catch serves the stale cache (marked "stale": true) when one exists, else 503 {"error":"pulse_unavailable"}. Success responses carry "status":"ok" so clients can distinguish authoritative data from legacy backends. NOTE: the edge function deploys out-of-band (supabase functions deploy community-pulse). - gstack-security-dashboard: captures the HTTP status; non-200 / network failure / error body / missing section → "unknown — backend error"; jq missing → "unknown — install jq" (the lossy grep fallback broke on nested arrays and under-reported attacks as zero — removed); a 200 without the new marker shows figures with an "unverified (legacy backend)" note. Also fixes a latent display bug: the TOTAL grep matched the digit 7 inside "attacks_last_7_days" and misreported every count. - gstack-community-dashboard: same class — curl || echo "{}" plus grep || echo "0" printed "Weekly active installs: 0" on any failure. Now "unknown — backend error (HTTP N)". test/security-dashboard-fallback.test.ts pins the matrix (200+marker, 200-legacy, 503, network failure) x (jq present, jq absent) for both bins: "unknown" states never render as 0. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(telemetry): redact error_message spans before they leave the machine (#1947) error_message was uploaded with only quote/newline escaping — stack traces and failed-API errors can embed credentials, private paths, and hostnames, and the sync path strips only _repo_slug/_branch. New lib/redact-engine.ts export redactFindingSpans(): replaces EVERY finding's span with <REDACTED-{id}> regardless of tier (applyRedactions is the interactive PII-only path and exits nonzero on credential findings, so it can't serve machine egress). Returns null when a span can't be located — callers drop the whole payload rather than risk a leak. gstack-telemetry-log pipes error_message through it at LOG time, so the local JSONL at rest is clean too; surrounding text survives for crash triage. FAIL CLOSED: bun missing, engine error, or non-JSON-string output all null the field. Tests pin: embedded ghp_ token → <REDACTED-github.pat> with context intact; redactor unavailable → null; raw bytes on disk never contain the token. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(redact): prepush guard fails closed on git failure; /ship owns hook install (#1946) Two gaps closed: 1. Fail closed. The git() helper returned "" on ANY non-zero exit or maxBuffer overflow (status null), addedLinesFor produced an empty string, and the push sailed through unscanned — fail-open on exactly the oversized-diff case where a large secret-bearing blob is most likely. The diff call now uses a strict variant that throws; main blocks with a clear message naming the GSTACK_REDACT_PREPUSH=skip escape valve. Probe calls (symbolic-ref, rev-parse, merge-base) keep the permissive helper — their failures are normal control flow. 2. Install path. The hook was installed by nothing ("opt-in, installed by nothing" was the issue's words). ./setup runs in the gstack checkout — the wrong repo for a per-project hook — so it gets a one-line hint only. /ship owns per-repo install: config redact_prepush_hook=true + hook missing → silent install (consent already given); config unset + no ~/.gstack/.redact-prepush-prompted marker → one-time machine-wide AskUserQuestion offer, answer persisted. ship/SKILL.md regenerated in this same commit (check-freshness bisect discipline). Tests: unscannable diff (bogus SHAs) → exit 1 + valve named; empty-but- successful diff → exit 0; static asserts pin setup as hint-only and the ship template as the installer surface. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * feat(redact): six new credential patterns — GitLab, HuggingFace, npm, DigitalOcean, Bearer, GCP SA (#1946) Coverage gaps from the #1946 security review, including token types for tooling gstack itself drives (glab): HIGH (block): gitlab.token (glpat-/glptt-/gldt-), huggingface.token (hf_), npm.token (npm_), digitalocean.token (dop_v1_), gcp.service_account (the JSON-escaped "private_key" form that dodges pem.private_key's literal-block match when minified, confirmed by "private_key_id" proximity). MEDIUM (warn): auth.bearer — the most FP-prone shape in the set (docs are full of "Authorization: Bearer <token>"), so it requires header-context proximity and the same entropy>=3.0 + placeholder validator recipe as env.kv. "Bearer YOUR_TOKEN_HERE" never fires; calibration over coverage, per the cries-wolf principle. All shapes are linear-time; test/redact-pattern-lint.test.ts covers them automatically. Engine tests add positive + placeholder-negative cases per pattern. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test: coverage-audit additions for the fix wave Ship Step 7 gap-fill (all passing, 248 tests across the touched suites): memory + dream stage probe-timeout proceeds, gbrain-detect override paths, stale-flag passthrough, 200-body-missing-.security fail-closed case, telemetry redaction edges, and credential-pattern edge cases. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: pre-landing review fixes Review army findings (1 critical, auto-fixed with regression tests): - CRITICAL (security specialist, verified live): redactFindingSpans spliced only the regex capture span, and pem.private_key / gcp.service_account capture just the BEGIN-header — the key body survived "redaction" and shipped via telemetry. Marker-only patterns now drop the whole payload (null, fail closed). Overlapping spans (Bearer+JWT on the same bytes) are coalesced before splicing so stale offsets can't leave partial secret bytes behind. - gitStrict: drop the dead `|| r.status === null` disjunct (null !== 0 already covers it); add the signal-kill/null-status regression test the docstring promised. - security-dashboard human mode flags stale snapshots ("figures may be out of date") instead of presenting frozen counts as current. - community-dashboard marker check uses jq when available — the grep-only variant misclassified whitespaced/reserialized bodies as legacy. - telemetry fail-closed test now shadows bun with a failing stub (deterministic on any host layout); stale "five status cases" describe title renamed. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix: adversarial review fixes (Claude + Codex cross-model passes) Both adversarial passes ran against the wave; every FIXABLE finding landed with a regression test: - probeTimeoutMs clamps to >=1ms: a fractional override floored to 0, and execFileSync treats timeout:0 as NO timeout — the probe that exists to bound hangs could hang forever (found by both models independently). - /ship silent hook install now requires the hooks dir to live inside .git: with core.hooksPath (husky's COMMITTED .husky/), the chaining installer would have renamed the team's committed pre-push and written a machine-local wrapper into the working tree (found by both models). - gstack-config gbrain-refresh accepts the "timeout" status — the last consumer still gating on literal "ok" (Codex); gstack-gbrain-detect's config-derived fields honor GBRAIN_HOME so the detection JSON can't report status ok alongside config_exists false (Codex). - prepush: a remote sha absent locally (shallow clone / stale fetch) falls back to the merge-base/empty-tree range — scans MORE, never blocks a legitimate push into training users toward --no-verify. - dashboards: curl's own 000 no longer doubles to "HTTP 000000"; the community dashboard flags stale snapshots like the security one; array sections parse via jq (the sed/grep loops truncated at the first ']'); the no-jq marker grep tolerates whitespace. - telemetry: multi-line redactor output nulls the field instead of corrupting the JSONL record; setup's hint fires only when the config key is genuinely unset (an explicit false is a recorded decline); the /ship prompt marker honors GSTACK_HOME. Kept as designed (cross-model tension noted): Bearer stays MEDIUM in the prepush gate — a HIGH Bearer would block every docs example; the entropy validator can't eliminate that FP class, and MEDIUM warns visibly. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore: bump version and changelog (v1.57.11.0) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * docs: P1 TODO — eval harness live progress + incremental persistence Root-caused during this ship: a killed eval run was indistinguishable from a healthy one for hours (per-file output buffering across mega test files, no incremental eval-store writes, no honest liveness signal). Full context and starting points in the entry. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test: fix operational-learning E2E fixture — copy lib/jsonl-store.ts Pre-existing breakage, proven on main: gstack-learnings-log has imported lib/jsonl-store.ts (shared injection patterns) since v1.57.5.0 / #1910, but the fixture copies only the bin scripts — the bin exits 1 before writing anything, on main silently (stderr swallowed) and on this branch loudly (the #1950 error-surfacing made the four-day-old failure visible). A real install always ships bin/ and lib/ together; the fixture now does too. Verified: the fixture-shaped invocation writes the learning (exit 0) with lib present, exits 1 on both main and this branch without it. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * fix(ios-qa): isolate E2E tests under --concurrent (3 real races) The ios-qa E2E file failed intermittently under `bun test --concurrent` (the eval harness default). Three distinct shared-state races, all fixed: 1. Shared pidfile: a module-level `workDir` reassigned in beforeEach was clobbered by parallel tests, so concurrent daemons collided on the same pidfile and the loser returned `already_running`. Each test now gets its own dir via makeWorkDir(). 2. process.env path globals: tests set GSTACK_IOS_AUDIT_PATH / _ATTEMPTS_PATH / _ALLOWLIST_PATH on the shared process env; concurrent tests stomped each other's audit/attempts destinations. Threaded auditPath/attemptsPath/allowlistPath through DaemonOptions (and mintForCaller) as explicit args — env is no longer load-bearing. 3. afterEach cleanup race: the per-test cleanup drained a shared dir array, so the first test to finish deleted still-running tests' workDirs mid-assertion. Moved to afterAll (cleans once, after all settle). Verified: 5/5 clean full-suite runs at --max-concurrency 15 (was intermittent); daemon unit suite 91/91; daemon source compiles. The paths default to the env-derived locations when options are omitted, so the production CLI path is unchanged. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * test(pty): pin spawned claude to EVALS model chain (default claude-sonnet-4-6) launchClaudePty spawned the interactive `claude` TUI with no --model flag, so the child inherited the operator's ~/.claude/settings.json model. On a slow-thinking model that meant 5+ min of extended thinking on empty plan-mode context, timing out the plan-mode smoke tests regardless of contention. Pin the model via opts.model ?? EVALS_MODEL ?? 'claude-sonnet-4-6' — byte-identical to session-runner.ts:144, so PTY and `claude -p` evals always agree. Pushed before extraArgs (last flag wins, so a per-test --model still overrides). Placement leaves the spawn region byte-stable for a clean merge with the in-flight hermetic-env branch. Plumbed model through the three plan-skill wrappers. Static-grep tripwires guard the pin, its fallback chain, the before-extraArgs ordering, and all three wrapper forwards. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(pty): detect markdown bold-bullet prose AUQs (fixes office-hours smoke) office-hours auto-mode renders its mode question as `- **Building a startup**` markdown bullets (office-hours/SKILL.md.tmpl:102) with no letter/number marker. isProseAUQVisible only matched `A)`-style lettered or `1.`-style numbered options, so the question went undetected: the model surfaced it at ~2m19s (well under the 300s budget) but the harness kept scoring the run "working" off the spinner glyphs and timed out — a false timeout on a question that was already on screen. Add Pattern 3: when an interrogative line ('?') is present AND 3+ bold-bullet markers (`- **`) appear in the 4KB tail, classify as a prose AUQ. Bold is the discriminator vs incidental prose bullets; the line anchor is dropped (stripAnsi can collapse option lines) and the existing `❯ 1.` cursor gate still defers to a live native list. Wires through the existing classifyVisible 'asked' path and the timeout high-water-mark, so office-hours now classifies 'asked' instead of 'timeout'. Five unit cases: the office-hours render passes; no-'?', <3-bullet, plain-bullet, and native-cursor cases stay false. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(pty): detect stripAnsi-collapsed prose AUQs + judge spinner-precedence The plan-eng/plan-design plan-mode + finding-floor smokes timed out even when the skill HAD rendered a complete prose AskUserQuestion and was waiting: the PTY strips cursor-positioning escapes, collapsing the option newlines/spaces so "A) ..." arrives as "A(recommended)" / "-B:" and "Reply with A, B, or C" as "ReplywithA,B,orC". Every line-anchored detector (Patterns 1-3) returns false on those bytes, so proseAUQEverObserved never latched and the run timed out on a question that was already on screen. Add Pattern 4/5: a two-signal collapsed-form detector — a reply/recommendation marker (space-insensitive "reply with [A-D]", "Recommendation:", or "(recommended)") AND 2+ distinct A-D letters each punctuated by ) : or (. The conjunction is what separates a real AUQ from incidental report prose; verified true on the verbatim failing-run buffers where Patterns 1-3 return false. Also fix the Haiku judge spinner bias: of 614 verdicts, 569 were 'working' and 95 of those noted a question was visible — Claude Code keeps the spinner animating at an idle prose decision, so the judge coin-flipped. Add a precedence override: when an option list AND a Recommendation/Reply instruction are both visible, classify WAITING even with spinner glyphs. Kept the strict dual-signal gate (never option-list-alone) so auto-decide-preserved doesn't flip. 5 unit tests pin the two-signal contract (2 true on real collapsed bytes, 3 false guards). 90 -> 95 pass. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(plan-review): ask-first scope gate for plan-eng + plan-design review On an empty/cold invocation, plan-eng-review and plan-design-review would dive straight into repo exploration (plan-eng) or a 7-pass mockup+audit (plan-design) and only ask the user much later, if at all. plan-ceo-review already asks first via an unconditional Step-0 gate and behaves well; these two did not. Add a hard-STOP scope gate as the FIRST operational instruction in each skill (above the design-doc check / pre-review audit / mockup defaults it explicitly overrides): the first tool call must be AskUserQuestion confirming the review target, before any git/Read/Grep/Glob/Bash or mockup generation. Under --disallowedTools the options render as plain column-0 lettered prose with a Recommendation + "Reply with A, B, or C" line so the answer is detectable. This is correct cold-start UX (confirm what to review before grinding a full review on nothing) and it is the product half of the plan-mode smoke fix; the harness collapsed-form detector is the deterministic half that catches the ask however it renders. Templates + regenerated SKILL.md (default variant). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(tiers): reclassify stochastic plan-eng/plan-design ask-first smokes as periodic plan-eng-review and plan-design-review run a long explore/audit before their first AskUserQuestion, so whether the plan-mode + finding-floor smokes reach a terminal outcome within the 300s/600s budget depends on stochastic ask-first compliance (measured ~50-67%/run even with the hardened gate). Per the "non-deterministic -> periodic" tiering rule, move the four affected smokes (plan-eng/plan-design review-plan-mode + finding-floor) to periodic. The deterministic harness fix (collapsed-form detector + judge precedence) and the ask-first gate lift these from always-failing to mostly-passing and are the real product+harness improvements; periodic monitoring tracks the rate weekly without blocking PRs on an LLM coin-flip. plan-ceo/plan-devex ask-first reliably and stay gate-tier. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * ci(evals): gate the deterministic PTY plan-mode smokes in CI The real-PTY plan-mode smokes never ran in CI — the gate was local-only. Add an e2e-pty-plan-smoke matrix suite running the two deterministically-reliable ones (office-hours-auto-mode, plan-mode-no-op) so a regression there blocks PRs. The stochastic plan-eng/plan-design ask-first smokes stay periodic (touchfiles E2E_TIERS) and are not CI-gated. A fresh CI container has no ~/.claude.json, so the spawned interactive `claude` would wedge on the onboarding + API-key-approval dialog. Add a scoped seed step (hasCompletedOnboarding + key approval, its own ANTHROPIC_API_KEY env) before the run — mirrors what the hermetic E2E child env seeds. Per-suite timeout override (35 min) via matrix.suite.timeout so the PTY suite has headroom for --retry 2 without bumping the other 12 suites. Report runner count 12 -> 13. Validate via workflow_dispatch before relying on the gate (PTY-in-CI is new). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * ci(evals): install gstack skill registry for the PTY smoke suite The first dry-run of e2e-pty-plan-smoke failed: the spawned interactive `claude` printed "Unknown command: /plan-ceo-review". .claude/skills is gitignored, so a fresh CI checkout has no gstack skill registry and the TUI can't resolve /office-hours or /plan-ceo-review. Add a Register step (scoped to the suite, after Seed, before Run) that mirrors setup's --no-prefix user-scoped registry minimally: $HOME/.claude/skills/gstack -> repo (resolves the preambles' absolute ~/.claude/skills/gstack/bin/* and <skill>/sections/* paths) + per-skill SKILL.md/sections symlinks for the two skills these tests invoke. HOME is /github/home in this container and the runner adds no HOME/CLAUDE_CONFIG_DIR override (no hermetic mode), so $HOME is the right anchor — the Seed step already proved claude reads it. No ./setup (binary build + Chromium + fonts + /dev/tty prompt); SKILL.md + bin/ + sections/ are committed. Self-validating: fails the step loudly on a dangling symlink or missing `name:` frontmatter, so a moved target surfaces here instead of as a silent 35-min "Unknown command" timeout. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v1.58.4.0) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
|
|
Garry Tan
|
c7ae63201a
|
v1.58.1.0 feat: hermetic local E2E + Conductor prose AskUserQuestion (#2004)
* feat: add shared call-time isConductor() helper
Single source of truth for Conductor host detection in TS consumers
(CONDUCTOR_WORKSPACE_PATH / CONDUCTOR_PORT). Reads the passed env at
call time, not a module-load snapshot, so unit tests can pin the env
inline without Bun --preload (esm-hoist-breaks-env-pin-bootstrap).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: harden question-preference-hook harness against ambient Conductor env
runHook copied all of process.env into the hook subprocess, so running the
suite inside Conductor (CONDUCTOR_WORKSPACE_PATH/PORT set) would leak those
markers. Strip them so the existing cases deterministically characterize
NON-Conductor behavior before the Conductor branch lands. Baseline: 15 pass.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: PreToolUse hook denies AskUserQuestion in Conductor, redirects to prose
Conductor disables native AskUserQuestion and routes through a flaky MCP
variant that returns '[Tool result missing due to internal error]'. The
hook now denies any AUQ call in a Conductor session and instructs the model
to render a prose decision brief instead (transport avoidance, not preference
enforcement) — firing for one-way doors too, with a typed-confirmation
requirement for destructive paths.
Precedence: never-ask auto-decide still wins (user already settled those);
Conductor prose is the fallback for everything else; non-Conductor behavior
is byte-for-byte unchanged. Restructured the per-question loop to compute
eligibility without early-returning so the Conductor branch can run as the
fallback while preserving memoryContext on every exit.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: Conductor renders AskUserQuestion decisions as prose by default
In Conductor, native AskUserQuestion is disabled and the MCP variant is
flaky, so skills now render every decision as a plain-text prose brief the
user answers by typing a letter — proactively, not as a failure reaction.
- Preamble emits CONDUCTOR_SESSION, gated on != headless so eval/CI inside
Conductor still BLOCKs instead of rendering prose to nobody.
- AskUserQuestion Format gains a Conductor-default-prose rule (auto-decide
preferences still apply first; prose decisions log via gstack-question-log
since PostToolUse never fires), a one-way/destructive typed-confirmation
rule, and a typed-reply continuation protocol for split chains.
- Regenerated all SKILL.md + ship golden fixtures; bumped affected carve
skeleton caps to absorb the always-loaded additions.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: deploy the Conductor AskUserQuestion hook (setup + upgrade migration)
The PreToolUse hook only delivers its Conductor-prose guarantee if it's
installed, but setup skips hook registration in non-interactive (conductor/CI)
setups. Two fixes so layer 3 actually deploys:
- setup: treat a Conductor workspace as an implicit opt-in for the PreToolUse
hook on the silent fall-through (never overriding an explicit opt-out).
- migration v1.58.0.0: re-register the hook for existing Conductor installs on
/gstack-upgrade, idempotent and respecting plan_tune_hooks=no.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: E2E for Conductor prose + fix auto-decide-preserved GSTACK_HOME bug
- New skill-e2e-conductor-prose (periodic): Conductor env + plan-eng-review
surfaces a prose decision brief, not a silent skip. Header documents this is
end-to-end behavior coverage; the deterministic Conductor guard is the
question-preference-hook unit test (the PTY harness can't register the MCP
variant — Codex #10).
- Fix the pre-existing bug in auto-decide-preserved: it seeded the never-ask
preference under GSTACK_HOME=tmpHome but never passed GSTACK_HOME into the
PTY run, so the spawned claude read the real ~/.gstack and the preference
was inert (Codex #9). Now passes GSTACK_HOME + CONDUCTOR_WORKSPACE_PATH to
prove auto-decide still wins over the Conductor prose redirect.
- Register both in touchfiles (periodic tier).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* v1.58.0.0 feat: Conductor renders AskUserQuestion decisions as prose
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: strip ambient Conductor env in memory-cache-injection hook harness
Same dev-in-Conductor leak fixed for question-preference-hook: this suite's
runHook copies process.env, so running it inside Conductor flipped the
defer-path memoryContext assertions into the [conductor] prose deny. Strip
CONDUCTOR_* so the cases characterize non-Conductor behavior. (CI is headless,
so this only bit local Conductor runs.)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: gstack-detach — run agent eval/bench jobs in their own session
Long agent-run jobs (30-60 min evals, benchmarks) die when the harness sends
SIGTERM to a background task's process group on turn boundaries / monitor
stops / interruptions (observed: 'script test:gate terminated by signal
SIGTERM'). gstack-detach runs the command in a fresh session (python3
os.setsid, or setsid on Linux, nohup fallback) so a group SIGTERM can't reach
it, and wraps it in caffeinate -i on macOS so idle-sleep can't kill it either.
Returns immediately; caller polls the logfile. Secrets stay in env, never argv.
The guard test pins the contract: the command runs in a different process
group than the caller and outlives the launching shell.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: eval:bg* scripts — detached eval runs for agents
Agent-facing convenience scripts that launch the eval suites through
gstack-detach so a harness SIGTERM can't kill a long run. eval:bg (diff-based),
eval:bg:all, eval:bg:gate, eval:bg:periodic — each returns immediately and
streams to /tmp/gstack-evals.log for polling. The plain test:evals / test:e2e
scripts stay foreground for humans.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: CLAUDE.md — agents must run long evals via gstack-detach
Codifies the detached-execution default: agent-launched eval/benchmark runs go
through bin/gstack-detach (or the eval:bg* scripts) so a harness SIGTERM or
macOS idle-sleep can't kill a 30-60 min run, then poll the log with a
death-aware watcher. Humans keep foreground scripts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: harden gstack-detach against all four eval-infra killers
The basic bash detach fixed SIGTERM but a real run on a shared dev box hit
three more killers: cross-worktree API saturation (15-way concurrency x a
sibling worktree mass-timed-out the suite), a silent hang (periodic bun died
with no exit marker), and shared-/tmp log contamination (a concurrent
worktree's agent output bled into the log). Rewrite as a portable python3 tool
that bakes in all four fixes:
- fork + setsid: SIGTERM-proof (own session, survives harness polite-quit)
- caffeinate -i on macOS: no idle-sleep death
- --lock NAME (fcntl, machine-wide): concurrent worktrees SERIALIZE instead of
saturating the shared model API
- run-scoped default log (~/.gstack-dev/eval-runs/<label>-<slug>-<branch>-<ts>-<pid>):
no cross-worktree collision/contamination
- --timeout watchdog + a guaranteed '### gstack-detach EXIT=<code> ###' sentinel
on every terminal path: no silent hang, finished-vs-died always detectable
Guard test pins all four: detached pgid differs + outlives launcher, run-scoped
log path, watchdog EXIT=timeout, and lock serialization (second run WAITS).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: eval:bg* use run-scoped logs + machine lock + watchdog
Drop the shared /tmp/gstack-evals.log path (the cross-worktree collision that
contaminated a live run) for gstack-detach's run-scoped default, and add the
machine-wide gstack-evals lock (concurrent worktrees serialize, no API
saturation) plus per-tier watchdog timeouts (60/90/120 min). Each eval:bg*
prints its run-scoped log path to poll.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: wire detached-eval guidance into /ship + correct CLAUDE.md flags
- /ship eval step (sections/tests.md): long eval suites launch via gstack-detach
(own session, machine lock, EXIT sentinel) so a turn boundary can't kill a
30+ min run mid-ship — the exact failure observed during this branch's ship.
- CLAUDE.md: correct the now-stale /tmp reference; document the --lock (serialize
worktrees, no API saturation), --timeout watchdog, run-scoped log, and the
guaranteed EXIT sentinel the poller breaks on.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor: extract pure promotedEnv() from conductor-env-shim
Single source of truth for GSTACK_* key promotion semantics. The ambient
promoteConductorEnv() becomes a wrapper; behavior-preserving. Needed by the
hermetic env builder which must not mutate process.env.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: hermetic child-env builder for E2E runners
Allowlist scrub (basics/network/named-auth kept; CONDUCTOR_*, CLAUDE_*,
GSTACK_*, MCP_*, GBRAIN_*, operator credentials dropped), per-runner
extraAllow, overrides merge last, EVALS_HERMETIC=0 byte-identical escape
hatch read at call time (ESM-hoist safe). Sync memoized singleton temp dirs
(<runRoot>/.claude keeps the extractPlanFilePath contract), seeded
.claude.json for non-interactive first run, pid-aware GC of crashed runs.
19 free unit tests.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: session-runner spawns hermetic children + isolation canaries
claude -p children now get the allowlist-scrubbed env and a gated
--strict-mcp-config (EVALS_HERMETIC=0 restores operator env AND args).
Two gate-tier canaries make the clean room falsifiable: hermetic-canary
asserts env redirect + scrub + zero MCP servers + nonzero API-key cost
from the Bash tool_result (never model prose); hermetic-sentinel plants a
poisoned operator config (user CLAUDE.md + MCP server) and proves the
child cannot see it. Empirically verified on claude 2.1.175: print mode
needs no seed config (the seed serves the PTY path); the child CLI sets
CLAUDECODE for its own tools, so that scrub is pinned in unit tests, not
E2E. hermetic-env.ts joins GLOBAL_TOUCHFILES.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: PTY runner spawns hermetic claude sessions
launchClaudePty children get the allowlist-scrubbed env, a gated
--strict-mcp-config, and the session exposes hermeticConfigDir for
forensics (hermetic plan files live under <dir>/plans/ and still match
extractPlanFilePath via the /.claude dir-name contract). Seeded trust
state covers repo-cwd sessions; the 15s trust-watcher stays as fallback.
Verified foreground via the plan-mode-no-op gate test.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: codex/gemini runners spawn hermetic children
Same allowlist scrub as the claude runners, with each provider's auth
surface re-admitted via extraAllow (codex: OPENAI_API_KEY/CODEX_* plus
its tempHome .codex copy; gemini: GEMINI_*/GOOGLE_* with real HOME for
~/.gemini auth). The gemini spawn previously inherited the full operator
env with no env property at all.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: agent-sdk-runner spawns hermetic children via complete Options.env
The historical 'env: breaks SDK auth' failure was partial-env replacement:
Options.env replaces the child's entire environment, so objects lacking
ANTHROPIC_API_KEY killed auth. Passing the complete hermetic env (key +
PATH + redirected CLAUDE_CONFIG_DIR/GSTACK_HOME) works — validated live
via query() with a Bash tool call (success, real cost, Conductor vars
scrubbed). Per-test opts.env merges last; ambient key mutation still
works because the builder reads process.env at call time.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: static tripwire pins hermetic wiring in all five runners
Free-tier invariants: every runner builds child env via hermeticChildEnv,
no raw ...process.env spread at any spawn site, --strict-mcp-config gated
on isHermeticEnabled in both claude runners, and no test callsite passes
the operator env into a runner's override parameter (scoped to runner
calls — unit tests spawning gstack bin scripts directly are exempt).
Mirrors the terminal-agent-pid-identity / server-embedder-terminal-port
tripwire idiom.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: refresh codex/factory ship goldens with detached-eval block
|
|
|
Garry Tan
|
14fc0866d9
|
v1.58.0.0 feat: diagram + multi-format document engine (mermaid, excalidraw, single-file HTML, DOCX) (#1990)
* docs(todos): P3 content-hash diagram render cache for make-pdf
Deferred from the diagram-engine eng review (Codex outside-voice D7):
repeat make-pdf runs re-render every fence; cache keyed on fence source +
bundle version once multi-diagram docs make it worth building.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(diagram-render): offline mermaid+excalidraw render bundle for browse
Single self-contained page (dist/diagram-render.html, 9.2MB, committed per
eng-review D2) exposing __renderMermaid / __mermaidToExcalidraw /
__excalidrawToSvg / __rasterize / __probeImage through browse load-html +
js --out. Render contract per D3: securityLevel strict, per-fence ids,
print-css font lock, htmlLabels off (canvas-taint-safe). Deterministic
build (same sha twice); drift test pins dist == BUILD_INFO == package.json
pins and rebuild-reproducibility when toolchain matches. Spike-proven
offline: flowchart + sequence SVG, editable .excalidraw scene, 300dpi PNG.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(diagram-render): __downscaleRaster for print-resolution image normalization
Data-URI rasters re-encode in their own format (JPEG stays JPEG at q0.9 —
PNG-encoding photos bloats them) at an explicit target pixel width. Used by
make-pdf's pre-pass for the 300dpi content-box ceiling (eng-review D4).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(make-pdf): diagram pre-pass — mermaid/excalidraw fences render as vector SVG; local images inline as data URIs
```mermaid / ```excalidraw fences extract to placeholder tokens, render in
one diagram-render bundle tab per run (reset contract: bundle page reloads
after any render error), and substitute back as accessible <figure> blocks
with the raw source preserved in a comment. Render failures produce a loud
red diagnostic block, never silent raw code. render=false keeps a fence as
code; title="..." becomes the aria-label and caption.
Local images now actually render: page.setContent loads at about:blank
(tab-session.ts:194), so relative paths silently 404'd before. The pre-pass
resolves them against the markdown's directory, inlines as data URIs, probes
intrinsic dimensions from the bytes (pure-TS PNG/JPEG/GIF/WebP/SVG sniffing),
and downscales rasters wider than 2x the content box at 300dpi. Remote URLs
warn (offline posture, --allow-network exempts); missing files get a visible
placeholder; --strict hard-fails both for CI pipelines.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf): diagram pre-pass unit suite + e2e render gates
34 unit tests (fence extraction incl. nested/tilde/unclosed/render=false,
info-string parsing, slot substitution, diagnostic/figure escaping + SVG
script strip, byte-level dimension probing across 5 formats, content-box
math, image inlining incl. strict/remote/missing/data-URI paths). E2E gate
proves through the compiled binary: both fences render as vector text
(id-collision check), raw mermaid ships only via render=false, broken fence
yields the diagnostic block, and the relative fixture image rasterizes to
colored pixels (CRITICAL regression for the about:blank image fix).
--strict exits non-zero on a missing image.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(make-pdf): width directives + conservative auto-landscape via CSS named pages
`{width=full|<pct>|<dim>}` and `{page=landscape|portrait}`
suffixes translate to data-gstack-* attrs in render() (before the sanitizer,
which keeps data- attributes; unrecognized brace groups stay visible text).
Default width rule needs no code: intrinsic CSS-px capped at the content box,
never upscaled — figure img max-width owns it.
Auto-landscape promotes a block to `@page wide { size: <pagesize> landscape }`
only when aspect >= 1.8 AND intrinsic width > 2.5x the content box (~1600px on
letter) AND diagram provenance (rendered fences) or a whole-word alt token
(diagram|architecture|flowchart|chart|graph) for plain images. {page=...}
forces or vetoes; fence info strings accept page=... too. preferCSSPageSize
is passed to Chromium only when a promotion exists, so every other document
prints exactly as before. False negatives are cheap; false positives feel
broken (eng-review P4, Codex challenge accepted).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf): width-policy unit suite + landscape e2e gate with negative fixtures
24 unit tests weighted toward the false-positive guards: wide screenshot
without an alt hint stays portrait, sub-threshold and tall images stay
portrait, deterministic 1560/1561px boundary, whole-word alt matching
('photographic' must not match 'graph'), page=portrait veto beats every
heuristic, diagnostic blocks never promote. E2E gate asserts pdfinfo
per-page boxes through the compiled binary: exactly 3 of 5 fixture blocks
get landscape pages (alt-hinted image, directive-forced image, wide sequence
diagram) while the unhinted screenshot and the veto'd diagram stay portrait —
plus the --toc combo proving TOC and named-page landscape coexist.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(make-pdf): --to html|docx output formats
--to html writes the assembled self-contained document directly (no print
round-trip): inline vector diagrams, data-URI images, zero network
references, plus an @media screen layer for browser reading. --to docx is
the content-fidelity export (eng-review P8): html-to-docx@1.8.0 (exact pin;
pure JS, bun-compile-verified) maps headings/tables/code/lists; diagrams and
SVG images rasterize at 300dpi of the content-box width via the render tab;
diagnostic figures convert to plain p/pre so the converter can't silently
drop an error. --format keeps its page-size-alias meaning; --to is the
output format, and the CLI says so when confused.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf): format gate — html no-network-refs + docx zip content checks
HTML: zero src/href network refs, no script/link tags, inline SVG diagrams,
data-URI images, screen layer, diagnostic survives. DOCX: valid OOXML zip
(document.xml + Content_Types), >=2 PNG media (diagram raster + fixture
image), headings + render=false source + diagnostic text in document.xml,
no leaked mermaid source from rendered fences. Plus --to validation UX.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(diagram): /diagram skill — English in, editable diagram triplet out
New skill: agent authors mermaid from the user's description and renders the
triplet through the offline diagram-render bundle in the browse daemon —
.mmd source (the single source of truth), editable .excalidraw (opens at
excalidraw.com, round-trips back through re-render), and SVG + PNG. Flowcharts
convert to fully editable scenes; other mermaid types render with an explicit
upstream-converter limitation note. Never ships an unrendered source file;
offline is the contract (no CDN fallback). Inventory rows in AGENTS.md +
docs/skills.md; generated SKILL.md + llms.txt via gen:skill-docs.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(diagram): paid E2E pair — gate triplet contract + periodic authoring judge
diagram-triplet (gate, deterministic functional): a fresh claude -p agent
following the skill extract must emit a parseable triplet — graph LR/TD in
.mmd, excalidraw scene with >3 elements, SVG markup, PNG magic bytes.
Verified live: pass, $0.17, 58s. diagram-authoring-quality (periodic,
LLM-judged): faithfulness/labels/size rubric with a diagnostic-path cap,
floor 6/10. Verified live: pass at exactly 6 with substantive critique.
Touchfiles select both on diagram/** and lib/diagram-render/** changes;
tier split per E2E_TIERS rules (eng-review D5).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(diagram): register /diagram in the skill coverage matrix
Gate: triplet contract + structural floor; periodic: authoring-quality judge.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(make-pdf): typography scale-up, zero image truncation, landscape vertical centering
Dogfooding round on the repo README surfaced four output-quality bugs:
- Type was too small everywhere: body 11→12pt, h1 22→26pt, h2 15→18pt,
cover title 32→56pt with poster spacing, cover meta 10→13pt, TOC 11→12pt
with tighter leading, code 9.5→10.5pt, tables 10→11pt.
- Zero image truncation, ever: the max-width cap was figure-scoped, but
markdown images render as <p><img> — a 1850px GitHub screenshot ran off
the page edge. Global img { max-width: 100%; height: auto; } cap.
- hyphens: auto put real 'dif-\nferent' breaks into the PDF text layer the
moment 12pt made lines wrap (combined-gate caught it). Clean copy-paste
is the product contract; left-aligned rag doesn't need hyphenation →
hyphens: manual.
- Promoted landscape blocks now vertically center. CSS flex/min-height
centering fragments into phantom empty landscape pages in Chromium
(bisected: min-height at ANY value; 3 promotions printed 5 pages), so
image-policy computes an inline margin-top from each block's known
aspect ratio against the landscape content box instead — fragmentation
handles margins fine. .page-wide also drops its explicit break-before/
after (the page-name change already breaks on both sides).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf): pin zero-truncation invariant, typography floor, centering math
Global img cap pinned as a regex invariant (the figure-scoped-cap regression
class); typography floor (12pt body, 56pt cover, 12pt TOC); .page-wide must
NOT carry min-height/flex (the phantom-landscape-page regression class);
centering margin math verified both ways (2400×1000 image → 1.38in,
2050×600 viewBox diagram → 1.93in, page-filling directive block → no margin).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: diagram + multi-format documentation across README, make-pdf skill, and how-to guide
README gains /make-pdf (Publisher) and /diagram (Diagram Maker) rows in the
sprint table. make-pdf's skill doc — the agent-facing contract — gains Core
patterns for mermaid/excalidraw fences (title/render=false/page= options),
the image policy ({width=}/{page=} directives, zero-truncation, conservative
auto-landscape), --to html|docx, and --strict, plus the --to vs --format
disambiguation in Common flags. New docs/howto-diagrams-and-formats.md is
the user-facing walkthrough: fences, directives, formats, /diagram triplet,
the mermaid racetrack trick, troubleshooting.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf): fill ship-audit coverage gaps — downscale, reset contract, excalidraw fence, WebP
Ship coverage audit found 9 gaps (85%); this fills the 2 HIGH + 3 MEDIUM and
most LOW. diagram-gate fixture gains a 4200px incompressible photo (the only
live coverage of __downscaleRaster AND the 64KB chunked jsViaBuffer eval
transport — asserted via the downscale stderr warning), an ```excalidraw
scene fence rendered through exportToSvg (vector labels + caption in
pdftotext, no leaked scene JSON), and the broken fence MOVED BETWEEN the two
mermaid fences so the second diagram rendering proves the D6.2 reset
contract end-to-end. New coverage-gaps.test.ts (16 tests): mock-tab reset
contract (exactly one reload, post-failure fence renders), excalidraw
fail-fast diagnostic without a bundle call, rasterize error fallbacks
(figure/tag kept, never silent), WebP VP8/VP8L/VP8X byte parsers,
landscapeContentBox a4/asymmetric margins, bare-token slot fallback,
resolveBundlePath env override + error shape, screenCss media scoping.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(make-pdf): pre-landing review wave — fence fidelity, injection hardening, Windows paths, transport rework
Review army (6 specialists + red team) findings, all fixed:
- Indented fences replay byte-for-byte and indented diagram fences are NOT
extracted (red-team conf-9: the pre-pass reconstructed fences at column 0,
splitting any list containing fenced code — every ordinary document).
- String.replace $-pattern injection killed at every seam: substituteSlots,
mergeStyle, img/src rewrites all use function replacements (a diagram label
containing $' duplicated the document tail).
- Big-expression transport reworked: browse `eval <file>` (one spawn, any
size, Windows-safe) replaces the 64KB chunked window-buffer eval — fixes
the per-chunk spawn cost, the char-vs-byte argv units, AND the Windows
32,767-char command-line ceiling in one move.
- Staged-bundle trust: content verified by hash even when the file exists,
and the rename-failure path re-hashes the survivor (sticky-bit /tmp EPERM
would otherwise ride a pre-planted file past the check).
- Windows drive-letter img srcs (C:/x.png) reach the local-path branch
instead of being swallowed as unknown URL schemes.
- DOCX rasterize-failure now embeds the decoded source as visible text —
returning the figure made diagrams vanish silently (converter drops svg).
- Fence source preserved as base64 data-gstack-source attribute (the comment
encoding corrupted every '-->' arrow); decodeFigureSource() round-trips.
- inlineLocalImages memoizes per path; file:// uses fileURLToPath; preview
prints a divergence note for fences/local images; --to docx strips the
watermark div and warns about print-only flags; TOC links resolve in
html/docx (heading ids assigned); waitForExpression sleeps instead of
busy-spinning; escapeHtml/svg-dims deduped to single definitions;
typography stragglers (blockquote 12pt, footnotes 10pt, 42em screen
measure); bundle BUILD_INFO gains srcSha256 for no-node_modules drift
detection; MAX_TARGET_PX shared guard.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: make-pdf gate covers the diagram-render bundle; bundle pinned to LF
make-pdf-gate.yml paths gain lib/diagram-render/** and the drift test (a
bundle-only PR previously skipped every render gate AND no CI lane ran the
drift check at all). .gitattributes pins dist html/json to LF so Windows
autocrlf can't break the hash-pinned bundle.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(make-pdf)+feat(diagram): review-wave test pins + skill transport hardening
Tests: indented-fence byte-for-byte replay + no-extraction-in-lists,
drive-letter local-path routing, $-pattern slot immunity, base64 source
round-trip ('A --> B' exact), existing-style merge preservation, DOCX
rasterize-failure surfaces source, srcSha256 + font-stack drift guards,
landscape veto asserted as some-portrait/no-landscape (layout-order-proof),
judge rubric cap lowered to 5 so it actually fails, vacuous error-shape test
removed honestly, tmpdir cleanup.
/diagram skill: base64 transport (template literals corrupted backticks/${
in sources), content-addressed staging with hash verification, and --tab-id
pinned on every browse call so a concurrent /qa session can't be clobbered.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(make-pdf): out-of-tree image reads warn; --strict makes them fatal (D8.1)
Local CLI semantics stay (absolute paths and ../ still inline, like pandoc),
but never silently: an agent PDF-ing untrusted markdown can't quietly embed a
file from outside the input directory into a shareable document without a
visible warning, and --strict pipelines hard-fail. Two unit tests. Also:
TODOS.md gains the deferred e2e-harness dedup entry (D8.2).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: pre-existing test failure in skill-e2e-bws operational-learning
Root cause was the fixture, not model behavior: gstack-learnings-log gained
an import of lib/jsonl-store.ts in the v1.57.5.0 injection-sanitization wave,
but the test copies only bin/ scripts into its sandbox — the inline bun
import failed and the script exited 1 before writing, on every run, on main
too (reproduced at
|
|
|
Garry Tan
|
a5833c413f
|
v1.57.10.0 feat: Codex review default-on across review/ship/plan/docs (#1966)
* feat(config): make codex_reviews the master switch for all Codex review Broaden the codex_reviews doc to describe it governing /review, /ship, /document-release, plan reviews, and /autoplan. Reject invalid values on set (preserving the existing value) so a typo can never silently flip paid Codex calls on or off. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(review): Codex review default-on across review/ship/plan/docs Add a shared codexPreflight() helper (constants.ts) that, in one bash block, reads codex_reviews, sources gstack-codex-probe, checks install + auth, and echoes a single canonical mode (ready/not_installed/not_authed/ disabled). All Codex resolvers route through it. - generateCodexPlanReview: opt-in question removed; the outside voice now runs automatically (default-on), falling back to a Claude subagent when Codex is missing/unauthed. Cross-model tension still gates on user approval (sovereignty preserved). - generateAdversarialStep: probe-based availability (install AND auth), distinct not-installed vs not-authed guidance; 200-line structured-review threshold unchanged. - generateCodexDocReview (new, wired via CODEX_DOC_REVIEW): reviews the release's docs against the shipped diff range, informational + an explicit apply-fixes decision point, never auto-edits. - autoplan Phase 0.5 now honors codex_reviews=disabled so the switch is truly global. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore(docs): regenerate SKILL docs + refresh ship golden Output of gen:skill-docs for the Codex-default-on resolver/template changes. Refreshes the factory-ship golden fixture (codex-host output unchanged — resolvers strip for the codex host). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(infra): widen size-budget guards for default-on Codex outside-voice The codexPreflight() block + CODEX_MODE branch prose (replacing the smaller opt-in question) grows plan-ceo/eng/devex-review and review by 5-7% over baseline. Each bump carries a comment justifying it as intentional capability, not slop. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: guard Codex default-on + config reject-on-set skill-validation: assert plan reviews no longer carry the opt-in question and render the default-on outside-voice, document-release carries the doc review, and the codex host strips all of it. gstack-config: codex_reviews defaults to enabled, accepts enabled/disabled, and rejects an invalid value while preserving the existing one. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(test): align gstack-config tests with defaults-fallback behavior Three tests (last touched v0.13.7.0) asserted get/list print empty for unset keys, but gstack-config falls back to the documented defaults table (get returns the default, list shows the active-values block). Update the assertions to the real behavior and split out an unknown-key case that does still return empty. Pre-existing red, unrelated to codex review. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * v1.57.10.0 feat: Codex review default-on across review/ship/plan/docs Codex cross-model review now runs by default on /review, /ship, all four plan reviews, /document-release, and /autoplan, governed by one master switch (codex_reviews, default enabled). Plan-review outside voice is default-on; /document-release gets a new Codex doc-vs-diff audit; every call site detects install AND auth and falls back to a Claude subagent with a clear reason. Disable everything with: gstack-config set codex_reviews disabled Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
8241949357
|
v1.57.9.0 feat: source-clean gbrain render (dev-setup --out-dir + machine-wide gbrain-refresh) (#1951)
* feat(gbrain-detect): add --is-ok live-detection exit-code gate Single source of truth for 'is gbrain usable'. Runs live detection (never reads the possibly-stale gbrain-detection.json) and exits 0 iff status is ok, so setup, bin/dev-setup, and gstack-config can gate brain-aware rendering on one shared check instead of re-grepping the JSON. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(gen-skill-docs): add --out-dir with surgical section-path rewrite --out-dir <abs-dir> mirrors the Claude skill tree (SKILL.md + sections) into a separate directory instead of writing in place, and rewrites the literal section-base path (~/.claude/skills/gstack/<skill>/sections/) in generated content to point at the out-dir. The rewrite is surgical: only /sections/ paths move; bin/, browse/, docs/ references stay pointed at the global install. Global extras (proactive-suggestions.json) are skipped in out-dir mode. Default (no flag) behavior is unchanged. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(dev-setup): render gbrain :user variant to an untracked workspace dir Stops the dev/Conductor workspace from dirtying tracked SKILL.md source. setup honors GSTACK_SKIP_GBRAIN_REGEN (passed inline by dev-setup, never exported) and skips the in-place :user regen; detection is still persisted (PID-unique tmp so concurrent workspaces can't clobber it). dev-setup instead renders the :user variant into .claude/gstack-rendered (gitignored, per-workspace) and repoints the workspace SKILL.md symlinks at it, so the workspace gets brain-aware blocks while the worktree stays canonical. dev-teardown removes the render. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(dev-skill): refresh the untracked brain-aware render on template change After the default in-place regen (which keeps the worktree canonical and runs validation), also re-render the :user variant into .claude/gstack-rendered when it exists, so live template edits reflect at the workspace's runtime. Never creates the render dir during plain template dev. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(gstack-config): gbrain-refresh renders brain-aware blocks into the install Extends gbrain-refresh to render the :user variant into the global install (~/.claude/skills/gstack) so every project's Claude sessions get brain-aware blocks, not just the gstack dev workspace. Guarded against mutating the wrong directory: the target must exist, not be a symlink (a symlinked install points at a dev worktree), and look like a real gstack clone (VERSION + package.json). Idempotent and self-documenting. CLAUDE.md's deploy section now notes that 'git reset --hard' reverts the blocks and to re-run gbrain-refresh. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: cover gstack-gbrain-detect --is-ok + dev-skill render refresh Fills the two automated-coverage gaps from the eng review: --is-ok exit-code gate (no-cli -> nonzero, healthy -> 0, plus an agrees-with-JSON no-skew check reusing the deterministic fake-gbrain harness) and a static tripwire that dev-skill re-renders the :user variant into the workspace render dir only when it already exists. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v1.57.9.0) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: document brain-aware dev-setup render for v1.57.9.0 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
1626d4857b
|
v1.57.7.0 feat: GSTACK REVIEW REPORT always declares unresolved decisions (#1916)
* fix(plan-devex-review): add missing gstack-review-log step
plan-devex-review carried the EXIT PLAN MODE GATE but never wrote a
review-log entry, so the gate's 'review log was called' check was
structurally unsatisfiable and the Review Readiness Dashboard / GSTACK
REVIEW REPORT had no plan-devex-review data to read. Add a Review Log
section before the dashboard read, logging the devex fields the report
parser already expects (status, scores, product_type, tthw, persona,
competitive_tier, unresolved, commit).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(review): make unresolved-decisions status mandatory in GSTACK REVIEW REPORT
The report's UNRESOLVED line was optional ('omit if empty') and the EXIT
PLAN MODE GATE only checked it 'if applicable', so a plan could ship with
no statement about open decisions at all — a missed ambiguity read
identically to a clean plan. Now every report ends with a mandatory
unresolved-decisions status as its final line: either the exact unbolded
sentinel 'NO UNRESOLVED DECISIONS', or a '**UNRESOLVED DECISIONS:**' block
of bullets. The gate blocks ExitPlanMode unless that final line is present.
generatePlanFileReviewReport: current-review items are listed from context;
prior reviews contribute an aggregate count computed as latest-fresh-row-
per-skill minus the current run (no double-count, dashboard 7-day window).
generateExitPlanModeGate: check #3 is now blocking with no 'if applicable'
escape; bolded sentinel does not satisfy it.
Tests: static guard in gen-skill-docs.test.ts asserts the mandatory status
across all six report consumers and the gate across gate-bearing skills;
skill-e2e-plan.test.ts asserts the written report's final line is the
status (and fixes a stale 'four review rows' -> five-row prompt).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(review): compress unresolved-status prose to fit parity budget
After merging origin/main (v1.57.3.0), plan-devex-review exceeded the 1.05x
parity ratio vs the v1.53.0.0 baseline. Rather than rebase the baseline,
compressed the new prose to stay under the cap honestly: the report's
unresolved-status block (~32 -> ~9 lines) and the EXIT PLAN MODE GATE's
final-line check (~7 -> ~5 lines), plus the plan-devex-review review-log
step. All load-bearing rules and the exact gate-checkable tokens are
preserved; the static guards in gen-skill-docs.test.ts still pass.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test: regenerate stale ship golden fixtures (#1909 follow-up)
#1909 (v1.57.3.0) added the always-loaded PR-title-version rule to ship's
template and committed the regenerated ship/SKILL.md, but did not refresh the
three ship golden fixtures, leaving the golden-file regression test red on
main. Regenerate them from current output. The diff is purely #1909 content:
the PR-title invariant line plus a previously-unresolved ${ctx.paths.binDir}
placeholder that current generation correctly resolves. No feature content
from this branch leaks into ship (ship does not consume the review report
resolvers).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(plan-devex-review): restore TIMESTAMP fill instruction in review-log
Adversarial review caught that compressing the devex review-log block dropped
the TIMESTAMP substitution guidance the three sibling plan-review skills carry.
A literal "timestamp":"TIMESTAMP" parses as JSON but is an unparseable date,
so the Review Readiness Dashboard's 7-day freshness window silently drops the
plan-devex-review row (and the report's prior-review aggregation loses it).
Restore the one-line instruction. Also: the plan-review-report E2E now derives
its last-line check from the report slice, not the whole file, so a mis-placed
report surfaces the real trailing content in the failure message.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(parity): rebase parity baseline v1.53.0.0 -> v1.57.7.0
The v1.53 anchor is four minor versions stale. v1.54-v1.57 (ship/plan carving,
carve-guards, AUQ prose fallback, the cross-session decision-log preamble) plus
this branch's mandatory unresolved-decisions status line pushed the three
plan-review skills past the 5% ratchet even after exhaustive compression. The
new baseline captures current UNION sizes (skeleton + sections/*.md, matching
what parity-harness measures) so the per-skill 1.05 ratio keeps catching future
bloat. The frozen v1.44.1 integrity anchor and the v1.47 size-budget baseline
are untouched.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: bump version and changelog (v1.57.7.0)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
|
|
Garry Tan
|
9cc41b7163
|
v1.57.6.0 fix wave: 8 community bugs (4 security guards failing open) (#1911)
* fix(ship): adversarial subagent no longer trips usage-policy denial on own security fixtures (#1899) The Claude adversarial subagent in /review and /ship was told to "think like an attacker" over the full diff. When the diff includes the repo's own security regression fixtures (real attack payloads, by design), reasoning adversarially over that material triggered Anthropic's real-time usage-policy safeguards and the subagent call was denied — blocking the review. Fix at the prompt's source of truth (scripts/resolvers/review.ts {{ADVERSARIAL_STEP}}): - Authorized-defensive-testing framing: declares this is the maintainer's own repo and that attack-pattern strings inside test/fixture paths are the project's own regression corpus to analyze, not material to expand on. - Fixture summary-mode diff: full content for non-fixture source, --stat/--name-status for test/fixture files, so raw exploit bytes aren't fed into adversarial reasoning. The subagent must state fixtures were reviewed in summary mode (no silent coverage cut). Reported by @bmajewski. Regenerated review/SKILL.md + ship/sections/adversarial.md. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(redact): detect modern sk-proj-/sk-svcacct-/sk-admin- OpenAI keys (#1868) openai.key (HIGH/block) used /\b(sk-(?:proj-)?[A-Za-z0-9]{32,})\b/, which stops at the first - or _ in the body. Modern OpenAI project/service-account/admin keys use base64url bodies containing - and _, so they never reached the 32-char run and produced ZERO findings — a HIGH credential failing open through /spec, /ship, /cso, and /document-*. Replace with explicit alternation, bare vs prefixed (not a globally-optional prefix, which would match malformed sk--... or separator-less sk-projabc...): sk-{proj,svcacct,admin}- + [A-Za-z0-9_-]{20,} | sk-[A-Za-z0-9]{32,} (legacy) Tests: the three previously-missed shapes now block; FP guards pin that hyphenated prose and malformed sk- strings do NOT match (HIGH tier blocks, so calibration matters). Reported by @jbetala7. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(redact): reject malformed --max-bytes instead of silently disabling the size guard (#1824) The oversize check is designed to fail CLOSED, but a malformed --max-bytes turned it fail-OPEN. bin/gstack-redact did parseInt(maxBytes,10) and passed it straight through; parseInt("foo") is NaN. The engine guarded with `opts.maxBytes ?? DEFAULT`, and ?? does not catch NaN, so `byteLen > NaN` was always false and the fail-closed block never fired. A negative value made `byteLen > -5` always true, blocking everything. Two layers: - bin/gstack-redact validates the RAW string (parseInt accepts "123abc"->123, "1.5"->1): require /^\d+$/ and > 0, else exit 1 with a clear message. - lib/redact-engine.ts hardens the fallback to Number.isFinite && > 0 else the default cap — a guardrail so the engine never silently runs uncapped even if a bad value reaches it directly. Tests: NaN and negative both fall back to the default cap (oversize still blocks); CLI rejects garbage/negative with exit 1. Reported by @jbetala7. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(learnings): cross-project trust gate is an allowlist, not a denylist (#1745) gstack-learnings-search --cross-project is documented as an allowlist — foreign learnings load only when user-stated/trusted, to stop one project's AI-generated learnings from injecting into another project's reviews. It was implemented as a denylist: `if (isCrossProject && e.trusted === false) continue`. Any row where `trusted` is missing/undefined (legacy rows from before the field existed, hand-edited rows, rows from other tools) passed `undefined === false` → false → admitted. Those rows leaked across projects. Flip to `e.trusted !== true`. Test: a foreign row with no `trusted` field is now excluded (true still included, false still excluded). Reported by @jbetala7. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(safety): one-way-door classifier catches "rotate ... password" (#1839) scripts/one-way-doors.ts is the secondary safety net for ad-hoc AskUserQuestion ids with no registry entry; a false negative auto-approves a destructive op. The revoke and reset credential patterns both include `password`, but the rotate pattern omitted it, so the most common phrasing ("rotate the database password") classified as a reversible two-way question. Add `password` to the rotate alternation so all three verbs are parallel. New test covers rotate+password, the revoke/reset/rotate parallel, and rotate's other nouns. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(review): route .mjs/.cjs/.mts/.cts changes to the backend reviewer (#1810) gstack-diff-scope backend detection matched only *.ts|*.js. Modern Node ships backend code as ESM (.mjs) / CommonJS (.cjs) and explicit-module TS (.mts/.cts); none matched any category, so a PR touching only those files reported no backend scope and the Review Army skipped the backend reviewer. Add the four module extensions to the backend case. Test covers all four. Reported by @jbetala7. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(brain-cache): loadMeta tolerates malformed _meta.json without crashing (#1879) loadMeta returned the parsed JSON verbatim. A valid JSON file that lacked the last_refresh map made three consumers (isStale, cmdInvalidate, refreshEntity) throw a TypeError dereferencing meta.last_refresh — the sibling last_attempt was already guarded, last_refresh wasn't. Fix in loadMeta: - Shape-guard: JSON.parse can return null/array/string/number; non-object → fresh meta. - Normalize ONLY the dereferenced maps (last_refresh, last_attempt). - Deliberately do NOT default schema_version/endpoint_hash. Leaving them absent makes schemaVersionMismatch()/endpointSwitched() force a rebuild (missing identity = mismatch = safe); defaulting them would suppress cache invalidation and trust a stale file of unknown provenance. Tests: missing last_refresh no longer throws; null/array/primitive treated as cold; missing schema_version forces rebuild instead of a trusted warm hit. Reported by @jbetala7. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(skills): anchor guard/freeze/careful hook paths so they survive CC 2.1.162 (#1871) The PreToolUse frontmatter hooks for guard, freeze, and careful invoked `bash ${CLAUDE_SKILL_DIR}/.../check-*.sh`. Claude Code 2.1.162 no longer populates ${CLAUDE_SKILL_DIR} in the skill-hook execution env, so it expanded to empty and every Edit/Write/Bash ran `bash /...` and errored — breaking the safety skills entirely. Frontmatter hooks run before any skill-body bash, so no runtime-resolved variable can fix this; the command must be a path that's valid at hook time. Anchor to the installed checkout: $HOME/.claude/skills/gstack/{careful,freeze}/bin/check-*.sh, where the scripts actually live. ($HOME is expanded by the hook shell.) Reported by @omariani-howdy. Regenerated the three SKILL.md from templates. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: v1.58.0.0 — fix-wave release notes, VERSION bump, #1882 TODO CHANGELOG entry for the 8-fix safety wave (#1899, #1868, #1824, #1745, #1839, #1810, #1879, #1871). VERSION + package.json to 1.58.0.0 (MINOR — coordinated multi-file safety fixes on top of main's 1.57.3.0). #1882 filed as the top TODOS.md item (scoped out of this wave per decision; host-config change touching all 52 skills, distinct from the #1871 hook fix). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(learnings): strip backticks from #1745 comment inside the bun -e block The #1745 trust-gate fix added an explanatory comment containing backticks (`=== false`) and the JS block is a double-quoted `bun -e "..."` bash string, so bash command-substituted the backtick contents on every cross-project search — polluting stderr with "command not found" and leaving a latent shell-injection / source-corruption surface in a security gate. Caught by the wave's own adversarial review (#1899 framing working as intended). Reworded the comments to avoid backticks and dollar-paren entirely; the gate logic is unchanged. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(golden): refresh ship golden baselines (#1899 prompt + main's PR-title line) The three ship golden fixtures were stale: main's v1.57.3.0 added the always-loaded PR-title invariant to ship/SKILL.md but did not regenerate the goldens (the golden regression test fails on main too), and the codex golden still carried an unresolved ${ctx.paths.binDir} token. Regenerated from the current generated ship skills, which also picks up this wave's #1899 adversarial-prompt framing (inlined for codex/factory). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
45cc95d5f4
|
v1.57.5.0 feat: cross-session decision memory + gbrain dream-stage call graph (#1910)
* feat(gbrain-sync): add cycleCompleted() cycle-state probe
Reads `gbrain doctor` cycle_freshness to classify whether a source has
completed a full cycle (completed/never/unknown). A fail naming this source
-> never; a fail naming only other sources -> completed; an absent or
unparseable check -> unknown, so an unrelated doctor failure never masks a
real state. Gates the automatic call-graph build on --full.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(gbrain-sync): --dream call-graph stage with lock-free gate + honest outcome guard
Adds a source-scoped `gbrain dream --source <id>` stage that builds this
worktree's call graph (code-callers/code-callees). Runs lock-free after the
sync lock releases so it never blocks sibling worktrees; a .dream-in-progress
marker dedupes concurrent dreams. --full auto-runs it only when the cycle was
never built; explicit --dream always forces; --no-dream opts out.
The stage parses the cycle's own output and reports the truth, not a flat
"built": a WARN when the schema pack can't extract code symbols, when the
embed phase failed for a missing key, or when 0 edges resolved; OK with the
resolved-edge count otherwise. gbrain exits 0 even when it skips on a held
cycle lock (e.g. autopilot), so that case reports SKIP, not success.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: ignore gbrain .sources/ local staging dir
gbrain writes per-source staging and capability-check artifacts under
.sources/ in the repo root. It's machine-local runtime state, not source.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* docs(gbrain): honest call-graph guidance in /sync-gbrain + pin works on gbrain>=0.41.38
sync-gbrain frames the --dream offer honestly: building a call graph requires a
code-aware schema pack, and the dream stage reports a WARN when it can't. The
verdict's Call graph row mirrors the dream stage's real outcome instead of
assuming a completed cycle means edges exist. The ## GBrain Search Guidance
block written into CLAUDE.md drops the old code-callers --source caveat:
gbrain >=0.41.38.0 honors the .gbrain-source pin for code-callers/code-callees.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(jsonl-store): shared audited JSONL plumbing (injection-reject + atomic append + tolerant read)
Single source of truth extracted for D2A: gstack-learnings-* and the upcoming
gstack-decision-* bins share one injection-pattern list, one atomic single-line
appender, and one tolerant reader. No more drift between stores.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(learnings-log): use shared hasInjection from lib/jsonl-store (D2A)
Replace the inline injection-pattern copy with the shared list. One audited
write-path rejection across learnings + the upcoming decision store. Behavior
unchanged (35/35 learnings tests green); learnings-search keeps its inline copy
because a structural test pins its bash/bun shape.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(decision): event-sourced decision-memory model (lib/gstack-decision)
decide/supersede/redact events on lib/jsonl-store; active set is computed (no
mutable status), dangling refs tolerated. Free-text is injection-checked and
redact-scanned on write (HIGH secret -> reject). Scope filter (repo/branch/issue)
for relevant resurfacing. File-only + reliable; gbrain not required.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(decision): bounded active snapshot + compaction (redact expunges, supersede archives)
writeSnapshot/readSnapshot/rebuildSnapshot give an O(active) bounded read for the
session-start hot path (D1A). compact() rewrites the log to active, archives
superseded decisions for history, and EXPUNGES redacted ones (dropped, never
archived) so an accidentally-captured secret leaves the store for good.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(decision): gstack-decision-log + gstack-decision-search bins (non-interactive)
Two bins mirroring gstack-learnings-* (D3A). log writes decide/--supersede/--redact/
--compact events + refreshes the bounded snapshot + enqueues for cross-machine sync;
search reads the O(active) snapshot, scope-filtered to current branch, newest-first,
--all to include superseded, --json for machines. Empty store returns silently
(no snapshot write on an empty read).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(memory): surface active decisions at session start + capture nudge (Context Recovery)
Context Recovery now shows recent scope-relevant active decisions (bounded read of
decisions.active.json via gstack-decision-search) and instructs the agent to treat
them as settled calls and to log durable decisions/reversals. Closes the Phase-1
capture->curate->resurface loop, reliable + file-only. Regen across all hosts folded
in (squash-with-regen); parity 10/10, freshness green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test: refresh ship golden baselines for the memory-loop preamble change
Context Recovery now emits the cross-session-decisions block, so ship's preamble
(all hosts) changed. Golden baselines are hand-maintained copies (gen does not
write them); refresh them from the fresh gen so golden-file regression passes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* docs(memory): document the cross-session decision-memory loop in CLAUDE.md
Adds a '## Cross-session decision memory' section: how to resurface
(gstack-decision-search) and capture (gstack-decision-log) durable decisions,
the supersede/redact/compact verbs, and a crisp durable-vs-trivial definition
so the store stays signal. Reliable file-only path; gbrain not required.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(memory): emit durable decisions from ship/ceo/eng/spec at structured points
Wires the four skills that finalize real decisions to capture them in the
cross-session decision store, from their STRUCTURED outputs (never free-text
scraping):
- ship: the version bump (level + why) at write time
- plan-ceo-review: accepted scope + verdict (branch-scoped)
- plan-eng-review: the architecture verdict + key call (branch-scoped)
- spec: the filed issue's core approach (issue-scoped)
All emits are non-interactive, schema-correct (content in decision/rationale,
source=skill, confidence 1-10), and best-effort (|| true) so a decision-log
failure never blocks the workflow. Includes regen across hosts + refreshed ship
golden baselines.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(memory): optional gbrain --semantic recall for decision search
Adds gstack-decision-search --semantic (with --query): appends a 'Related from
memory' block from gbrain semantic search, scoped to the curated-memory source.
Pure enhancement, reliability-first: a new lib/gstack-decision-semantic.ts is the
ONLY decision module that touches gbrain and is imported lazily only on --semantic,
so the reliable file path never loads gbrain code. Every path degrades to the
reliable file results when gbrain is off, unconfigured, empty, or errors (never
throws, 10s timeout).
Built against the verified gbrain 0.42.x surface (text output [score] slug --
snippet, NOT JSON; curated-memory source resolved by worktree path, not a
gstack-brain-<user> id). Deterministic-contract tests only: parser units,
degrade-to-null when gbrain absent, and a fake-gbrain shim proving scope+search
end-to-end. find-contradictions deferred (no verifiable CLI surface yet + curated
memory not indexed).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(gbrain-sync): self-heal stale autopilot lock (dead-pid)
detectAutopilot treated a lock FILE as proof of life, so a crashed gbrain daemon
left a stale lock that wedged every sync forever (observed: a dead pid refused
--full indefinitely). Now read the holder pid (bare or JSON body) and check
liveness via signal-0: ESRCH=dead → ignore the stale signal and keep checking;
EPERM=alive (other user) → active. A stale lock never masks a live autopilot
process. Pure decision function — does not delete the file; the caller may clean it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* docs(review): drop stray trailing code fence in TODOS-format
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(test): align section-loading E2E testNames with their TOUCHFILES keys
Pre-existing on main (v1.56.x): the two section-loading E2E tests used
human-label testNames ('/ship section-loading') that don't match their slug
keys ('ship-section-loading') in E2E_TOUCHFILES/E2E_TIERS. Every other E2E test
uses the slug as its testName, and the TOUCHFILES completeness gate requires
testName to be a registered key — so the gate was red. Align both testNames to
their slug keys (also fixes tier lookup for these two periodic tests).
Verified failing on a clean origin/main checkout before the fix.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix: pre-landing review fixes (datamark, DRY, compact, coverage)
Addresses the pre-landing review findings (all INFORMATIONAL, no criticals):
- security: datamark resurfaced decision text at the render boundary
(lib/gstack-decision.ts datamark() — neutralizes code fences, --- banners,
<|role|>/</system> markers, control chars, newlines). Applied in
gstack-decision-search human output so stored text can't masquerade as
instructions in Context Recovery (codex hardening #3 / AC #7). --json stays raw.
- DRY: extract resolveSlug/gitBranch/flagValue to lib/bin-context.ts; both
decision bins use it instead of duplicating the helpers.
- compact(): batch the archive append (one write, not N) and shrink the
mid-compact crash window; simplify the opaque branch/issue ternary.
- coverage: learnings-log injection rejection (D2A wiring), search --recent/
--scope + NaN-safe --recent, datamark-applied, unparseable lock body,
compact-empty, corrupt-snapshot degrade.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(security): close adversarial-review findings in decision memory
Adversarial review (Claude subagent) found a CRITICAL the specialist pass missed:
- F1 (CRITICAL): 'Human:'/'Assistant:' turn-prefixes bypassed BOTH the write-time
denylist AND datamark(), landing verbatim in agent context inside the trusted
ACTIVE DECISIONS fence. Add 'human:' (+ 'disregard previous', 'from now on') to
the shared denylist, and have datamark() neutralize Human:/Assistant:/System:/User:
turn-prefixes (ZWSP) at the render boundary.
- F2: datamark() only stripped ASCII C0; extend to Unicode line terminators
(U+0085/2028/2029) and U+007F so 'strip newlines' actually holds.
- F3: validateDecide blocked only HIGH secrets; MEDIUM-tier PII (e.g. SSN) persisted
silently and synced cross-machine. The store is non-interactive (no confirm path),
so fail closed on MEDIUM too.
- F4: compact() was a lock-free read-modify-rewrite that could clobber a concurrent
append (lost decision). Add an O_EXCL compact lock + a pre-rename size recheck that
aborts untouched (skipped=true) if an append landed; caller re-runs.
- F7: filterByScope unknown/garbage scope fell through to 'return true' (leaked into
every context); fail conservative (false).
F5 (pid reuse) and F6 (pgrep over-match) are intentionally left as-is: both fail SAFE
(over-refuse sync); making them precise would introduce a fail-DANGEROUS path
(allowing sync during a real autopilot). True disambiguation needs gbrain to stamp the
lock with a start-time, which gstack doesn't own. F8 (compact moves history to archive)
is by design.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(security): close cross-model (Codex) adversarial findings
Codex adversarial review found a HIGH the Claude pass missed plus 3 mediums:
- C1 (HIGH): gstack-decision-search --all returned every decide and IGNORED redact
events, so a redacted secret still resurfaced via --all until compact ran. --all
now excludes redacted (redact = expunge from every read path), still showing
superseded history.
- C-med: semantic (external gbrain) slug/snippet were printed raw — datamark them too
so a gbrain hit can't spoof role markers / fences into agent context.
- C4: semanticRecall fell back to an UNSCOPED gbrain search when no curated-memory
source resolved, pulling code/doc corpora mislabeled as 'related decisions'. Now
returns null (degrade) when there's no worktree-backed memory source.
- C5: validateDecide scanned only decision/rationale/alternatives; branch and issue
are stored + surfaced (raw via --json), so include them in the injection+secret scan.
C2 (snapshot staleness) / C3 (compact TOCTOU residual): accepted for a single-user
store — atomic appends never lose the event, rebuilds self-heal, and the compact
size-recheck leaves only a sub-ms window; full append-locking would break the
lock-free append design.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: bump version and changelog (v1.57.5.0)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
|
|
Garry Tan
|
41c6d3ebf6
|
v1.57.4.0 refactor(ethos): rename Boil the Lake principle to Boil the Ocean (#1912)
* refactor(ethos): rename Boil the Lake principle to Boil the Ocean Reframes the completeness principle so the ocean (the complete thing) is the goal and lakes are the boilable units you ship on the way there. "Don't boil the ocean" was right when engineering time was the bottleneck; AI killed that bottleneck, so the ocean is now the destination. Resolves an existing split: the scope_appetite psychographic, archetypes, and the completeness intro flow already used "boil the ocean" as the complete-implementation pole while the named principle still said "lake". Sources only: ETHOS.md philosophy, CLAUDE.md, README.md, the preamble resolvers, and the plan/autoplan/document-generate templates. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: update assertions + golden fixtures for Boil the Ocean rename skill-validation and terse-build now assert "Boil the Ocean"; the three ship golden fixtures are regenerated to match the renamed Completeness Principle header and intro prose. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: regenerate SKILL.md files for Boil the Ocean rename Mechanical `bun run gen:skill-docs` output: the Completeness Principle header and intro flow now read "Boil the Ocean" across every generated skill. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v1.57.4.0) Boil the Ocean rename: completeness principle renamed across ETHOS, every generated skill, CLAUDE.md, README, and the preamble resolvers. Text only, no runtime behavior change. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
d8c91c6267
|
v1.57.3.0 fix(ship): always-loaded PR-title-version rule + fork-PR title-sync backstop (#1909)
* fix(ship): restore always-loaded PR-title-version invariant to skeleton
The v1.54.0.0 carve moved the 'PR title MUST start with v$NEW_VERSION' rule
out of the always-loaded ship skeleton and entirely into the lazily-loaded
pr-body.md section. The agent only set the version prefix if it happened to
read that section before creating the PR, so PRs landed with bare titles.
Restore a one-line invariant (+ helper reference) to ship/SKILL.md.tmpl right
before the {{SECTION:pr-body}} pointer, mirroring the AUQ always-loaded
precedent. Full procedure stays sectioned. Regenerated all hosts.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(ship): guard PR-title-version rule + pull_request_target safety
Two free gate tests so a future carve or workflow refactor can't silently
regress:
- ship-pr-title-version-always-loaded: asserts the invariant lives in the
always-loaded ship/SKILL.md skeleton (not only sections/), and that the
skeleton+sections union keeps BOTH the create and the existing-PR update
title paths. Modeled on test/auq-format-always-loaded.test.ts.
- pr-title-sync-workflow-safety: static tripwire that fails CI if
pr-title-sync.yml checks out PR-head code or inlines an attacker-controlled
${{ github.event.pull_request.* }} field inside a run: block (the two
pull_request_target footguns actionlint cannot catch).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ci): pr-title-sync covers fork PRs via hardened pull_request_target
Under plain pull_request the GITHUB_TOKEN is read-only on fork PRs, so the
title-sync backstop could never edit a fork/agent PR title. Switch to
pull_request_target (write token in base context) and make it safe:
- Check out the base repo only (no ref:) — execute trusted infra, never
fork-head code.
- All attacker-controlled PR fields (title, head repo, head sha) pass via
env: and are referenced as shell-quoted "$VAR", never inlined into run:.
- Read the PR-head VERSION as data (raw media type) from the head repo at the
head sha; guard the assignment under set -e.
- Same-repo read failure fails loudly; fork miss warns and skips (the backstop
stays green without going silently optional).
- Never echo the raw fork title (Actions parses ::workflow-command:: from stdout).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(ship): expand binDir path in pr-body Linked Spec block
ship/sections/pr-body.md.tmpl:98-99 used ${ctx.paths.binDir}, but the
gen-skill-docs generator only resolves {{TOKEN}} syntax in .tmpl files — the
${...} JS-template-literal form is substituted only inside .ts resolver files.
So the token passed through literally into the generated pr-body.md, leaving the
agent with an unexpandable ${ctx.paths.binDir}/gstack-paths command in the
Linked Spec auto-detect block. Use the hardcoded helper path, consistent with
every other path reference in this section.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(test): fold ship PR-title skeleton guard into carve-guard registry
main shipped a generalized carve-guard system (PR #1907) that is now the single
source of truth for carved-skill skeleton invariants. Register the PR-title rule
there instead of a standalone test: ship's mustStayInSkeleton asserts v$NEW_VERSION
+ the rewrite helper stay always-loaded, and mustMoveToSection asserts both the
create and update PR paths stay carved into pr-body.md (present in the union, out of
the skeleton). Delete the standalone ship-pr-title-version-always-loaded test it
replaces. The CI-workflow safety tripwire stays standalone (not a carve concern).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: bump version and changelog (v1.57.3.0)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
|
|
Garry Tan
|
4dfdb7cdc2
|
v1.57.2.0 feat: AskUserQuestion prose fallback when the tool fails at runtime (#1908)
* feat(auq): add gstack-session-kind + echo SESSION_KIND in preamble
Classifies the session as spawned | headless | interactive from env markers
(OPENCLAW_SESSION / GSTACK_HEADLESS / CONDUCTOR_* / CLAUDE_CODE_ENTRYPOINT / CI),
defaulting to interactive. Echoed once at skill start alongside BRANCH/REPO_MODE
so the AskUserQuestion-failure fallback can branch without a shell-out at failure
time. Degrade-safe: empty/error => interactive.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(auq): prose fallback when AskUserQuestion fails (interactive sessions)
On a genuine AUQ failure (tool absent, or present-but-erroring like Conductor's
flaky MCP returning '[Tool result missing due to internal error]'): retry once,
then branch on SESSION_KIND — spawned auto-chooses, headless BLOCKs, interactive
renders a prose decision brief the user answers by typing a letter.
The prose fallback MUST surface the triad: a clear ELI10 of the issue, a
per-choice Completeness score, and a recommendation+why (one paragraph per
choice). Carves out the [plan-tune auto-decide] denial as NOT a failure, and
qualifies the former 'tool_use, not prose' assertions so the rule isn't
self-contradicting. Tests pin the triad, the SESSION_KIND branch, the OV2
collision guard, the always-loaded guarantee, and a cross-file invariant on the
auto-decide prefix.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(auq): default GSTACK_HEADLESS=1 in eval/E2E runners
Headless harness runs classify as headless (BLOCK on AUQ failure rather than
emit a prose question no one reads). SDK runner uses ambient mutation, not the
Options.env object, to avoid breaking the SDK auth pipeline. Interactive-path
suites opt out by overriding the env per-run.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(auq): defensive PostToolUse error-fallback hook (OV3:B)
When an AskUserQuestion call returns an error/missing result, this hook injects
additionalContext reminding the model to run the prose fallback for the current
SESSION_KIND. It does not render prose itself — it guarantees the reminder fires
at the moment of failure instead of relying on the model recalling SESSION_KIND.
Inert on success and inert if the platform never invokes PostToolUse on tool
errors (unverified — could not force the Conductor MCP error in a harness; see
the spike doc). The prompt-level fallback covers the case regardless. Decision
logic is unit-tested deterministically; registered in setup beside the existing
AUQ hooks.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore(auq): regenerate SKILL.md for all hosts + refresh ship goldens
Regenerated from the resolver changes (gen:skill-docs --host all). Refreshes the
byte-exact ship golden fixtures (claude/codex/factory). Spec prose tightened so
the cross-cutting preamble addition stays under the 5% per-skill parity ceiling
(investigate 4.8%) — guard unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(test): kebab testNames for section-loading E2Es to match TOUCHFILES keys
The two section-loading E2E tests used display-form testNames ('/ship
section-loading', '/plan-ceo-review section-loading') while every other E2E
testName and their E2E_TOUCHFILES keys are kebab. The completeness gate does an
exact `name in E2E_TOUCHFILES` check, so it failed (pre-existing on main); diff-
based selection also couldn't match them. Align to ship-section-loading /
plan-ceo-section-loading.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(test): make external-host freshness checks deterministic
The parameterized host smoke + --host all freshness tests assumed an external
`gen:skill-docs --host all` had run first (it never does in `bun test`), so which
host reported STALE varied by sibling-test timing — flaky. Regenerate the
gitignored external host dirs in a beforeAll so the --dry-run check is
deterministic. It still catches non-deterministic generation (the real bug class
for regenerated outputs); the tracked-claude freshness test runs earlier and is
unaffected.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(parity): headroom for AUQ cross-cutting addition on carved document-release
Merging main brought the carve of document-release (smaller skeleton); the AUQ
prose-fallback adds ~2KB to every skill's always-loaded preamble, landing
document-release at ~5.9% over the pre-carve v1.53.0.0 baseline. Add a per-carve
maxSizeRatio override (CARVE_GUARDS single source of truth) and bump only this
skill to 1.08. All other skills keep the strict 1.05 ceiling.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(auq): harden error-fallback hook + harness per adversarial review
Codex pre-landing review found three real issues:
- The PostToolUse fallback hook shared source 'plan-tune-cathedral' with the
question-log hook (same event+matcher); gstack-settings-hook replaces the entry,
so it would have clobbered plan-tune capture. Give it its own 'auq-error-fallback'
source (separate entry, both run); ALREADY_INSTALLED now requires both sources.
- isErrorResponse triggered on any string containing 'internal error'/'is_error',
so a real answer or a {"is_error": false} payload could fire the fallback after a
successful question. Narrow it to the missing-result sentinel + boolean is_error.
- The SDK runner mutated process.env.GSTACK_HEADLESS process-wide (leaked headless
into later tests). Removed; GSTACK_HEADLESS=1 now lives in the eval package.json
scripts, scoped to the invocation and inherited by the SDK child.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: bump version and changelog (v1.57.2.0)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
|
|
Garry Tan
|
e722c5bf89
|
v1.57.0.0 feat: carve-guard system + carve cso/document-release/design-consultation (#1907)
* test: canonical CARVE_GUARDS registry; derive parity + size-budget from it Single source of truth for the carved-skill set + per-skill invariants (EQ1). parity-harness.ts sectioned entries and skill-size-budget.ts SECTIONS_EXTRACTED now derive from it instead of hand-maintained lists. Closes a pre-existing drift: plan-devex-review was in SECTIONS_EXTRACTED but had no sectioned parity invariant; now generated. carve-guards.ts is a pure leaf data module (import type only) to avoid an import cycle. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: shared carve-guard check fns with injectable root discoverCarvedSkills/checkOrdering/checkCompleteness take a root param so the negative tests can point the real guards at a fixture dir. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: E2 data-driven carve static ordering guard (gate) Per-PR backstop for every carved skill, one test() per skill, driven by CARVE_GUARDS staticInvariants. Generalizes + retires the ceo-specific ordering test. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: E1 carve-guard completeness meta-guard (gate) Asserts filesystem carved set == CARVE_GUARDS set both directions, so a future carve without a registry entry fails CI. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: ET1 guard-of-guards negative tests (gate) Temp fixture broken 3 ways proves E1/E2 actually throw, via the injectable root. Kills the silent-pass-guard failure class. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: T2 data-driven behavioral section-loading guard (periodic) One file iterating CARVE_GUARDS, one test() per skill with GSTACK_CARVE_SKILL cost-scoping (D-CODEX A). external carves (ship, plan-ceo) keep bespoke tests; testNames aligned to their touchfile keys. Registered in touchfiles. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: defer E3 real-session carve canary to TODOS Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat: carve document-release into skeleton + on-demand section Steps 2-9 (per-file audit, auto-updates, risky-change asks, CHANGELOG voice polish, cross-doc consistency, TODOS cleanup, VERSION bump, commit + PR body) move to sections/release-body.md, read on demand after the Step 1.5 coverage map. Skeleton 59,256 -> 45,797 B (-23%); union preserved. Adds the CARVE_GUARDS entry (auto-extends parity + size-budget via EQ1). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat: carve design-consultation into skeleton + on-demand section Phases 3-6 (complete proposal, drill-downs, design preview, writing DESIGN.md) move to sections/proposal-and-preview.md, read on demand after product context + research. Skeleton 80,719 -> 59,229 B (-27%); union preserved. Adds the CARVE_GUARDS entry. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat: carve cso into skeleton + on-demand section (security-safe) Scope-dependent audit Phases 2-11 move to sections/audit-phases.md. Mode dispatch (## Arguments, ## Mode Resolution), always-run Phases 0/1, and the Phase 12 false-positive-filtering exceptions stay ALWAYS-LOADED in the skeleton. Skeleton 79,383 -> 65,117 B (-18%); union preserved. Adds a cso CARVE_GUARDS entry with an earliest-use invariant (mustPrecedeStop): mode dispatch must appear before any STOP-Read, so a directive that decides which sections to read can't be stranded behind the STOP that reads them (codex outside-voice #6). carve-guard-checks gains the mustPrecedeStop check. parity moves cso monolith -> generated carved entry. cso-preserved.test.ts strengthened: phrases checked against the union, plus an always-loaded contract on the skeleton (dispatch + FP-filtering, codex #5). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test: make redaction/taxonomy tests union-aware for cso + document-release carves The cso carve moved Secrets Archaeology (prefixes, lib/redact-patterns.ts pointer, git-history scan) into sections/audit-phases.md, and the document-release carve moved the Step 9 PR-body redaction scan into sections/release-body.md. Three content-presence tests asserted that content in the skeleton SKILL.md/.md.tmpl; they now read the skeleton+sections union (same fix as cso-preserved + parity). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: bump version and changelog (v1.57.0.0) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix: address pre-landing review (codex) on the carve - cso section: add a scope-gate header so '--owasp' (and other scoped modes) run only their selected phases, not every phase bundled in the section ('execute in full' no longer overrides Mode Resolution). - carve-guard-checks: gateAfterStop now compares against the LAST STOP, not the first, so a gate stranded between two STOPs in a multi-STOP skeleton fails. - TODOS: behavioral section-loading hermeticity (verifier matches global-install path, not the fixture) — pre-existing in auq-sdk-capture.ts, deferred. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
|
|
Garry Tan
|
476b0ec597
|
v1.56.1.0 fix(sync): staging-dir ownership guard + resume-correctness fixes (#1802) (#1856)
* fix(sync): fail-closed staging-dir ownership guard — prevent rm -rf of repo (#1802) Adopts community fix #1827 by @diazMelgarejo (cyre). New lib/staging-guard.ts exports checkOwnedStagingDir(), the single fail-closed predicate for 'safe to recurse-delete or resume into', wired at cleanupStagingDir() (the deletion chokepoint), decideResume(), the ingest entry point, and makeStagingDir() (mints the .gstack-staging marker). Fixes #1802. Co-Authored-By: cyre <diazMelgarejo@users.noreply.github.com> Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(sync): don't route the remote-http persistent transcript dir through cleanup (#1802) The ingest finally ran cleanupStagingDir() unconditionally, but in remote-http mode stagingDir is the PERSISTENT transcript dir (~/.gstack/transcripts/) that gstack-brain-sync push must consume. The remote-http branch documents the intent to skip cleanup, but a finally runs on its return. Gate the call on !remoteHttpMode so the ownership guard only ever sees .staging-ingest-* dirs. Pre-gate this dir was deleted outright (broken artifacts handoff); post-#1827 it produced a false 'prevent data loss' warning every sync. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(sync): preserve staging dir on internal import timeout (#1802 C3) The import-timeout branch printed 'checkpoint preserved' but the finally then deleted the staging dir: the SIGTERM forwarder's preserve branch only runs when the PARENT is signalled, and an internal runGbrainImport timeout kills just the child and returns normally. So #1611 resume-after-timeout never actually worked. Mirror the forwarder in the timeout branch: set preserveStaging only when gbrain checkpointed against this dir (finally then skips cleanup); otherwise clean up and tell the user it restages instead of falsely promising a resume. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(sync): resume must not mark failed files as ingested (#1802 C4) On resume, stagedPathToSource was rebuilt as an empty Map, so readNewFailures() could not map gbrain's per-file failures back to source paths. Every failure fell through to state recording — failed files were silently marked ingested and never retried. Reconstruct the map from the prepared pages via a shared stagedRelPath() helper (single source of truth with writeStaged, so the keys can never drift). Exports stagedRelPath + readNewFailures for a behavioral test proving the reconstructed map recovers the failure the empty map dropped. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * harden(sync): close staging-guard TOCTOU + fail hard on marker write (#1802 C5) checkOwnedStagingDir() now returns the realpath-resolved canonicalPath on a pass, and cleanupStagingDir() rmSync's that instead of the raw input — closing the gap where the input is a symlink swapped between the ownership check and the delete. makeStagingDir() tears down the partial dir and rethrows if the marker write fails, so a marker-less dir (which the guard would refuse forever) can never leak. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore: v1.56.1.0 — staging-dir ownership guard + resume-correctness fixes (#1802) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * ci: grant the eval report job issues:write so PR comment upsert stops 401ing Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: cyre <diazMelgarejo@users.noreply.github.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |