* perf(gen-skill-docs): stop emitting the "When to invoke" body section
The Claude catalog trim moved each skill's routing prose ("Use when...",
"Proactively invoke...", voice triggers) out of the frontmatter description
into a "## When to invoke this skill" section at the top of the body. Claude
Code reads only name + description to choose a skill and loads the body after
the skill is invoked, so that text cannot route anything and costs tokens on
every invocation. Keep the one-line trimmed description; drop the section.
The routing prose stays in each template's description, which
--catalog-mode=full and the other hosts still render. Tests that pinned the
section now assert its absence, or check the template instead.
(cherry picked from commit 707ccd6f38)
* feat(gen-skill-docs): keep voice triggers in the root routing block
The Claude catalog trim drops each skill's "Voice triggers (speech-to-text
aliases)" line from its description, and the body section that used to hold
it is gone. {{VOICE_TRIGGER_ROUTING}} renders them in the root router's
routing block instead: one "Voice trigger ... → invoke /<skill>" line per
skill, read from the templates' voice-triggers lists, so disabled-skill
filtering applies to them like any other routing line. Other hosts keep the
line in each description and render nothing here.
extractVoiceTriggers moves to the new resolver so the generator and the
resolver share one parser.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 645e75b1e8)
* chore: regenerate SKILL.md files without the "When to invoke" section
Output of `bun run gen:skill-docs --host all` after the previous commits.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit ff4c813337)
* test: refresh context-budget ceilings for the smaller skill bodies
`bun test/helpers/capture-context-budget.ts`. Against a recapture of the base
commit, 55 skills that lost the section get a lower ceiling and the root
gstack router rises by 532 tokens for the voice-trigger routing lines (5,369
tokens lower in total). devex-review and eval-plan differ from the committed
fixture only because it already lagged main. alwaysOnTotal is unchanged: the
catalog descriptions did not change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit d886fcecc1)
* test: refresh the claude ship golden without the "When to invoke" section
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit f1789dca5c)
* test: lower the claude design-review size pin
The size guard (#3199) asks for it: claude/design-review shrank by 437 chars
without its "When to invoke" section.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 3a446fcc62)
* test: skip alias symlinks in the voice-routing check, keep unrelated budget ceilings, add #3129 CHANGELOG entry
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* test: rebuild the historical cookie approval with the dropped section, record the skill-fixture change as non-behavioral
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* v1.91.157.0 perf(gen-skill-docs): drop the "When to invoke" body section from Claude SKILL.md files (port of #3129)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
---------
Co-authored-by: thanhnguyentung95 <17350411+thanhnguyentung95@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* docs: approved reliability follow-ups plan (Oct 2026)
* D1: score overlay wrong answers as comparison results (contract v4)
A completed answer that fails the fixture's output check is now a valid
measurement with taskCorrect false and answerError set, in either arm.
Only execution, scope and metric failures invalidate a trial. ON
correctness still gates the case, so an OFF wrong answer stays in the
comparison. OVERLAY_CONTRACT moves to v4; v3 verdicts are not
reinterpreted. Replays the 37198445662 sonnet trials (OFF-0 wrong).
* C4: score auq-matrix substance with a 3-sample judge panel
Each skill keeps one native capture in the same execution; the
recommendation-substance grader now runs as a JUDGE_PANEL_SAMPLES panel on
that capture and gates on the mean against the unchanged minimum of 4. A
failed sample fails the panel without resampling. Approved at the gate
2026-10-04 for censuses after this change.
* E1: one Bun version source; setup refuses below the 1.3.3 floor, updater checks before advancing
bin/gstack-bun-version.sh holds the security floor (1.3.3) and the tested
version (1.4.0, the CI pin and engines.bun). setup refuses Bun below the
floor before writing anything, warns between floor and tested, warns and
continues on an unparseable version, and prints a 1.4.0 install hint.
Prereleases sort below their release (semver).
gstack-session-update now fetches, reads the incoming release's floor and
checks the bun setup will run before fast-forwarding to that exact
revision; below it the update is held with reason bun-too-old and a
resume command, leaving the revision and installs untouched.
Fixture copies of setup (team-mode, claude-code migration) now include the
sourced file; the slow-git lock test sleeps on fetch, the updater's
network step.
* E3: record the owned Chromium's PID over CDP so the #2709 reap runs on Playwright 1.62
Playwright 1.62's Browser has no process(), so chromiumProcInfo was always
null and browse stop never reaped a surviving headless Chromium. The owned
headless launch now reads the browser PID from CDP SystemInfo.getProcessInfo
(bounded at 2s; a hung or failed call logs once and records nothing). The
reaper's start-time and cmdline identity checks are unchanged, so a recycled
PID or an externally owned browser is never signalled.
* A1: ship docs gate records post-child hashes with gstack-docs-candidate, saves each record once
Stored transcripts (37198445662 gate-census-2) show the parent's reads of
release files take ~1 s; the parent's own model time dominates (completion
268 s of 392 s, store ~270 s of 387 s), and 45-78 s of it is retyping hash
maps, child output and records into several artifacts after the child
returns. The child contract is unchanged.
* A2: /plan-ceo-review sends the mode handoff line before any other tool call
plan-ceo-mode-routing's HOLD SCOPE session spends its 240 s post-selection
window on model work: 37176837432 skipped the handoff helper and paraphrased
the mode, 37198445662 ran it, then wrote the plan, loaded sections and probed
Codex for 114 s before the posture chat. The handoff line is the posture.
* B3: accept wrapped unchanged rows in the cropped Edit preview
PR run 37176835584 timed out with the floor's owned Edit menu on screen:
unchanged diff rows soft-wrap under their gutter with a blank marker
column, and a trimmed empty unchanged row keeps only its number. The
parser accepted wrapped +/- rows only, so it never verified the owned
edit. Both forms are now parsed and still compared with the exact source
lines; the captured viewport is replayed with negative controls.
* A3: record per-phase headroom diagnosis for docsync, plan-mode-no-op, mode-routing and gate slice 11
* E4: /review Step 1 keeps going when the base fetch fails and reports stale base coverage
Under Codex's read-only .git, 'git fetch && merge-base && diff --stat'
stopped at the fetch, so the empty output read as 'Nothing to review'.
The fetch is now separate, prints BASE_REFRESH fresh or stale with the
local base revision, and the review states stale coverage.
* D2: reword the format-approach question prompt
Both plan-ceo-review-format-approach refusals in census 37195203538
(Fable 5.1 model_refusal_no_fallback, category reasoning_extraction:
"duplicating model outputs") were anchored on the session prompt, whose
capture instruction asked for the "verbatim" question and "the exact
prose you would have shown" in a "format-capture test". The approach
case now asks for the same question in the same file as the step's
output, without that wording. The other three plan-format cases keep
their unchanged prompt (no refusals in 72 trials).
* C2: /qa defines its outcome filename fields; QA report templates name the measured timing fields
Low judge samples in failing panels named both: qa workflow (37186854666)
'an undefined {user} in the outcome filename', and qa-only workflow
(37198445662) 'template asks for Duration / durationMs totals while
reporting.md mandates Probe budget and Guarded command time'. The qa
union stays under its 1.11 size ratio (93292 of 93294 bytes).
* 0.1: session ledger written by the claude -p, Agent SDK, PTY and Codex runners
Each armed model session appends one best-effort line to
$GSTACK_EVAL_DIR/session-ledger.jsonl: a stable session key, the timeout it
armed, monotonic elapsed time, its end reason (completed, session_timeout,
observer_timeout, api_error, refusal, aborted, error), one sanitized evidence
line and, for stream-JSON runners, a compact liveness summary (longest
in-request silence with nothing outstanding, last event kind, tool open at
end). Partial messages now stream for every claude -p session and for SDK
sessions without a custom query provider (runSharedInteractive opts in); their
rows feed only the summary, so transcripts keep their size unless
publicStreamDiagnostics is set. Structured refusal and API-error signals win
over phrases, and tool results are never read for them.
* 0.2: failure_cause beside failure_class, from runner facts and the session ledger
One pure failureCauseOf() beside failureClassOf() sets failure_cause
(contract > pre_turn_infra > api_error > refusal > provider_stall >
session_timeout > observer_timeout > assertion > unknown; no machine-set
'detector') with one bounded evidence line, and failureDetailOf() keeps a
Bun matcher's Expected/Received or a judge's failing dimensions in
failure_detail. Both record builders (classifyTrialShard, junitCensus) call
them and attach the trial's ledger sessions. failure_class, error and
TRIAL_ERROR_MAX are computed exactly as before; the new fields are optional,
bounded and validated only when present, and a frozen copy of the v1 reader
accepts every new record. A provider stall needs partial messages and
STALL_WINDOW_MS (120 s, eval-budgets.ts beside HEADROOM_WARN/FAIL) of
in-request silence with nothing outstanding; it stays a failure under
EVAL_POLICY v1.
* E4: fingerprint under a read-only .git via a private object dir owned by review-log
Codex's sandbox mounts .git read-only, so gstack-wtree's git add -A could not
write blobs and /review lost its start capture. gstack-review-log now creates
a private 0700 object directory under TMPDIR, passes it to gstack-wtree (which
writes new objects there with the common-dir object store, resolved before
exporting, as an alternate, so linked worktrees work), exposes it to the
bindReview/checkSharedLibsReuse consumers that read the tree and blobs, and
removes it on EXIT/INT/TERM after they finish. gstack-wtree prints one stderr
line and exits 1 on a failed write or unwritable TMPDIR instead of printing a
partial hash; review-log no longer swallows that line. Standalone gstack-wtree
keeps writing to the repository store so git diff between saved snapshots
(/ship) still works.
* 0.3: red lines name their values, evidence and an after-a-repair command; one published-text sanitizer
Each red line in the census report keeps its stored error and adds the
machine cause with its evidence, the failure_detail values (Expected/Received,
or each failing judge dimension with its mean, sample count and the sample
sentence naming it), the slice artifact and shard directory that hold the
transcript, a fetch command (eval:pass-rates --run <id> --case <id>), and
'after a repair:' with a command that selects only that case (E2E ids via
--case, judges via their selection and an exact test-name filter). A
session that completed and failed a check says so instead of 'assertion
(success)'. Report text that reaches a PR comment, issue or committed file
goes through scripts/lib/published-text.ts: control and bidi characters
stripped, credential-scan patterns redacted, mentions neutralized, HTML and
link brackets escaped inline, blocks in a fence longer than any backtick run,
length capped. Verdicts and collector verdict objects are unchanged.
* E2: one make-pdf resolver for readiness and every $P block on env-var hosts
make-pdf exported $P from its readiness block, which Codex and the other
env-var hosts lose between blocks. binaryAssignment gains 'make-pdf': the
MAKE-PDF SETUP readiness check and the runtime prelude both use it, so every
block that runs $P derives the same make-pdf/dist/pdf binary, with
MAKE_PDF_BIN (dev builds) winning when it is executable. Claude output keeps
its repo-local-first probe; only the false 'subsequent blocks' comment
changed. Root plus P=... fits the 400-byte prelude budget; the fixed
MAKE_PDF_OVERRIDE segment is the one exemption, named in runtime-root.ts and
the conformance test.
* 0.1: eval:pass-rates --headroom, --reds and --run report views; 85% headroom alarm in --gate
A new pure module, scripts/lib/eval-history.ts, reports per case the slowest
armed session against the timeout it armed (censored when it timed out,
insufficient below 3 samples, unknown without a session ledger, with the
case wall shown separately as an upper bound), the census critical path from
the slice jobs' timings, the verdict red ledger per census by failure class
and machine cause with shrunk per-case rates, and the all-green probability
(an approximation: the product of 1 - p_i, with an interval). --run <id>
lists one census's reds with their values and history and downloads only the
slice artifacts those reds name, reporting any over 64 MB instead of skipping
them. eval-flake-rank.ts only routes flags: --help works offline, an unknown
flag or case id exits 2 before any fetch, --case narrows the JSON and the
alarms, every view prints its run/branch scope, and --gate also alarms on a
session above HEADROOM_FAIL (85%) of its budget.
* E5: GSTACK_DESIGN_IMAGE_MODEL override; the live model check always tests the defaults
The image_generation tool model gains an env override validated against a
conservative gpt-image-<version> pattern; a bad value fails before any
request and the error names the variable. Pairing and rejection hints name
the tool model actually requested. design/scripts/live-model-check.ts now
ignores GSTACK_DESIGN_MODEL and GSTACK_DESIGN_IMAGE_MODEL (and says so), so
an override cannot mask a retirement of a default, and its exit 2 (no key)
says nothing was checked.
* C3: /deslop-shared-libs names a concrete risk for each recommendation, never 'none'
Both shared-libs-opportunity-judgment reds (37193478719 t3, 37195203538 t1)
failed adoption_and_risk because the top recommendation said 'Main risk:
none found inside the repo' / 'none in behavior'. The skill asked to 'name
the main risk or uncertainty' without saying what counts.
* B4: --case selects loop-registered cases; an unknown id exits before any process
caseSelection() resolves the one paid file that registers a case and how a
trial selects it: by its literal Bun test name, by the id a registration loop
names its test at runtime (coverage audits, TPA, journeys), or by running the
whole file when it registers no other case (plan-mode-no-op, whose test names
are not the id, so the old name pattern ran nothing). A whole-file case runs
one file shard per trial and is judged by shard status, as CI judges it.
--list prints the selection. Every E2E_TIERS id now resolves except
carve-section-loading, which fifteen files register (CI runs each as its own
shard) and which still fails with the owning files named.
* 0.3: after-a-repair runs the shard file when --case cannot resolve a case
A case that several files register (carve-section-loading) cannot be selected
by --case; its red line now prints the bun test command for the shard file
that went red, so every printed command selects exactly one case.
* C1: pre-register workflow-judge control bundles for qa, qa-only, review and setup-deploy
Positive (production bundle), negative (headings only) and one targeted
execution-blocking defect per judge, one 3-sample panel each, built with the
eval's own reader, prompt builder, options and floors.
* F: cost_known: a trial whose harness captured no billing says so; totals never hide it
cost_usd stays a finite number. Trial and eval records gain an optional
cost_known boolean (absent means known, so old records still parse; the
validator accepts only a boolean). Session-ledger rows record whether the
session's terminal result carried total_cost_usd; PTY and Codex sessions
never do, and SDK captures without a terminal result do not. A trial is
cost-known only when it has eval records and every record and session
captured billing, so JUnit-only PTY cases (125 of 298 trials in run
37198445662 recorded 0) now read cost unknown. The census headline prints the
known sum plus the unknown count, never a bare total over unknown costs, and
no total when every cost is unknown; the history import keeps the flag.
PTY transcripts are not priced: test/helpers/pricing.ts estimates exclude
cache writes, so they cannot give the billed cost.
* B1: replay corpora for the five repeat-offender cases
shared-libs-plan-callers, plan-ceo-section-loading, auto-decide-preserved,
plan-design-review-plan-mode and ship-docsync-late-result each get an
inventory of every wave census in scope (all artifacts still available)
and minimal, redacted detector or actor inputs replayed through the real
detector in the free suite: census passes, every distinct census red, and
derived negative controls labelled as derived. Fixed false reds stay as
expected passes with their history; the two genuine auto-decide product
failures (37176837432, 37179171083) stay refused. A contract test bounds
each entry at 64 KB, scans it for HIGH credentials, requires a pass and a
refusal per case, and reports entries whose series identity is stale.
Building the auto-decide corpus exposed a latent native detector misread:
a bold 'Mode application and rationale.' heading was parsed as a
withdrawn Mode field (37186854666). A field status now ends at its
sentence.
* B2: read approved stale-fill decisions from more valid ledger records
The structured ledger reader is authoritative before the prose fallback,
but it missed two valid approved records in passing censuses
(37186854666: a camelCase writeProfile subject; 37195203538: a
slash-joined 'D1 / STALE-FILL' row id), which then rested on the prose
detector that caused both historical reds. Both forms are accepted; an
unresolved row, an unrelated decision, the copied fixture plan and
non-write camelCase words stay refused.
* C1: record workflow-judge control results (negatives separated; qa and review miss a removed step)
* C2: record every low workflow-judge sample in the wave censuses with its disposition
* 0.3: published lines cannot break a fixed fence
evals.yml wraps collector-outcomes failures in a fixed three-backtick fence
for the PR comment, so those lines (and eval:pass-rates output meant for the
tracking issue) go through sanitizeFixedFenceLine: cleaned like every
published line, with backticks turned into quotes. Inline published text
also neutralizes backticks.
* A1: fit the post-child record wording under ship's 1.428 parity size cap
* E4: inline the Step 1 base refresh to fit review's 1.199 parity size cap
* B2: inventory of gate and periodic phrase-only detector checks
docs/evals/detector-inventory.md lists every pass/fail decision in gate
and periodic E2E cases that rests on phrases in model free text, with
its structured source (if any), whether the contract is user-visible,
and a disposition: already structured-first, convert candidate, phrase
witness required, or no structured source. Convert candidates are
ordered by census red history.
* 0.2: failureCauseOf takes the stall window instead of importing it; the PTY ledger never throws
eval-store.ts imported STALL_WINDOW_MS from eval-budgets.ts, which free tests
replace with mock.module stubs (review-finalization-budget, office-hours,
qa-fix-loop, qa-only), so loading eval-store failed there. failureCauseOf is
now dependency-free: the record builders pass STALL_WINDOW_MS. A PTY session
whose handle cannot show its text (test doubles) records no panel instead of
throwing from runPtySession's cleanup. The 37186854666 replay also runs
through classifyTrialShard.
* A2/auto-decide: the CEO handoff chat repeats the helper's printed line verbatim
Claude Code 2.1.284 collapses the helper's Bash output ('Ran 1 shell
command'), so the 'Auto-decided review mode → …' line reaches the user only
through chat. In lane D's auto-decide-preserved corpus every passing capture's
chat starts with that line, and the reds paraphrased it ('HOLD SCOPE was
auto-selected…', 'Mode decision done…', "I'm defaulting to HOLD SCOPE…").
The template now says verbatim and why.
* C2: record the judge panels rerun on the final ship and review wording
* integrate: touchfiles and module-size ratchet entries for the wave's new modules
GLOBAL_TOUCHFILES gains session-ledger.ts, published-text.ts and
auq-substance-panel.ts; bin/gstack-bun-version.sh joins every entry that
lists setup or the updater; auq-matrix stays a rule case with its C4
evidence beside E2E_KINDS; eval-history, published-text and session-ledger
are registered new modules.
* E1: BROWSE SETUP install hint pins the tested Bun (1.4.0)
* E2: quote "$P" in make-pdf commands so install paths with spaces work
* E1: /gstack-upgrade checks the incoming Bun floor before pulling
gstack_bun_incoming_hold (bin/gstack-bun-version.sh) is now the one
pre-advance check: the auto-updater calls it before its fast-forward and
/gstack-upgrade's git path calls it after fetching. Below the incoming
release's floor the upgrade stops with BUN_TOO_OLD and the checkout,
VERSION and installs stay untouched. Vendored installs were already safe:
setup refuses before writing and the swap restores the previous install.
* 0.3: only main posts or closes tracking issues; reports carry alarms, red ledger, headroom and all-green
A branch-dispatched periodic or marathon run writes its report to the step
summary and a report artifact; the upsert and close steps run only on
refs/heads/main. The periodic body now keeps every pass-rate alarm line (the
old grep kept only the header), adds the red ledger and session headroom read
with an explicit --branch, the all-green estimate and the triage guide link,
and routes report text through scripts/lib/published-text.ts. The marathon
lane gains the matching close-on-green step.
Migrated: evals-workflow-wiring pinned report-summary.md in the upsert step;
the body is now written by 'Write the census report'.
* C3: prosons neutral-posture detector ignores a negated "taste call"
Both plan-review-prosons-neutral-neg reds (37179171083 t3, 37193478719 t3)
recommended A and closed with "a coverage call, not a taste call"; the
old /taste call/i matched the negation. The detector moves to a helper so
the two census captures replay free as negatives, with positive controls.
* B2: credit the CEO mode handoff only when the chat shows its line
Claude Code 2.1.284 collapses the gstack-ceo-mode-handoff Bash result, so
the printed AUTO_DECIDE line is user-visible only when a later chat message
of the session begins with it. The native detector now requires that
witness; the helper's tool result locates it but no longer replaces it, and
a generic mode statement cannot stand in once a handoff ran. Census
37182865432 is relabelled a genuine miss (expected: refused): the chat said
"I'm defaulting to HOLD SCOPE…" and "Mode decided: HOLD SCOPE". Corpus
replay: 5 credited, 3 refused. ceo-mode-handoff tests migrated to the
witness rule (deliberate behavior change).
* test: baseline the wave's approved detector and harness replay tests
B1 corpora, B3, C3, C4 and D2 replay captures through paid-case detectors,
actors or fixtures (test code by design); each entry carries its reason.
* E6: live Codex multi-block periodic case in the host-run Codex job
test/codex-e2e-multiblock-live.test.ts runs ./setup --host codex once from the
job's checkout into a fresh HOME and CODEX_HOME (auth.json copied from the CI
login home), asserts codex debug prompt-input lists the gstack router as a
real file, seeds one learning, and runs /learn's Show recent block through
codex exec: that later block must root itself through its prelude, run the
installed gstack-learnings-search and print the seeded learning; no block may
report a missing install or an unresolved gstack helper path.
The file joins CODEX_CI_FILES so the planner gives it to the host-run Codex
job. paid-test-durations.json carries a 180 s packing estimate (refreshed by
--write-durations after the first census) so the plan keeps every slice
within max-parallel. Oracles live in test/helpers/codex-multiblock.ts with
free replays in test/codex-multiblock.test.ts. Registration (E2E_TOUCHFILES,
E2E_TIERS periodic, E2E_KINDS rule) is the integrator's.
* E5: design live-model check as a periodic case
test/skill-e2e-design-model-smoke.test.ts runs design/scripts/live-model-check.ts
(default models only) under a temp HOME. Without OPENAI_API_KEY the case is a
Bun skip that the census lists as SKIPPED with no coverage credit; with a key,
exit 0 passes and any other exit fails with the script output (exit 2 under a
present key fails as 'nothing was checked'). paid-test-durations.json carries
a 90 s packing estimate. Registration (E2E_TOUCHFILES, E2E_TIERS periodic,
E2E_KINDS rule) is the integrator's.
* A1: docsync parent points children at a saved interface file
The /ship docsync fixture saved its observation interface only in the
parent prompt and told the parent to copy it into every child prompt,
about 30-40 s of parent generation per session (lane S, 37198445662).
docsSessionOptions now writes it to <home>/fixture-interface.md (0600) and
the parent tells each child to Read that file first. Fault-actor prompts,
which prepare already writes to a file, are unchanged. Fixture-only.
* A3: a JUnit case's recorded wall is capped by its shard's wall
junitCensus summed a case's JUnit test times as duration_ms; for
plan-mode-no-op's concurrent PTY tests that is 346 s inside a 176 s shard
(the "432-613 s" history). The case wall headroom shows is now
min(sum, shard elapsed), a true upper bound. Session headroom already
used the ledger's per-session clocks.
* A3: refresh paid-test durations from census 37198445662
Recorded 2026-09-29 estimates under-packed gate slice 11 (17 files
estimated 1078 s ran 1353 s; plan-review-report 76 s -> 198 s,
plan-ceo-plan-mode 37 s -> 112 s, safety-codex-consult 83 s -> 150 s).
Refreshed both tiers with --report --write-durations from the latest
census's slice walls. Old plan re-costed with observed walls: worst slice
~677 s (slice 11); refreshed plan: every slice estimated <= 539 s.
* C2: Browser Quick gets a feasible budget; QA loop names its directory and stop path
Judge rationales on the qa-only panels (2 of the last 3) cited the 30 s
Browser Quick budget: homepage + 5 targets, each a checkpoint write and a
guarded command, left ~5 s per probe. Quick is now 180 s (method text: 3
minutes); the qa-only-no-fix paid check expects the same native deadline.
PROBE_DIR points at its source line, a stopped loop goes to the final
report steps (annotations, materialize, Markdown), and /review and /ship
say that after smoke expiry only plan checks rerun in revalidation.
* C2: name the flows judges found implicit in review resolvers
Structured outside review: MISSING COVERAGE asks no fix question and is
never reported clean (was "preserve the existing user decision flow").
/review's HIGH-impact plan gate: spawned or non-interactive runs report
REQUIREMENTS MISSING and continue, writing no TODOs. Value cards: the byte
clamp applies to card lines; JSON records you write keep full values.
* C1: a free test catches a removed workflow step, which the judges missed
C1 controls showed the qa workflow judge passing /qa without Phase 10 and
the review judge passing /review without Step 3. The new test pins every
step/phase heading of qa, qa-only, review, ship, plan-ceo-review,
plan-eng-review and setup-deploy (derived from the templates' step
headings) and names any that goes missing or out of order; new step
headings must be pinned. The judge rubric is unchanged (EVAL_POLICY).
* size: fit the wave's wording under ship, review and qa parity caps
Measured after all lane merges: ship -79, review -210, qa -68 bytes over.
Trimmed the added wording first (post-expiry rule, MISSING COVERAGE,
plan-gate default, value-card JSON, stop line) and folded /review's
redundant INFORMATIONAL sentence and findings bullet. ship 268,036 (room
8), review 130,118 (room 1). qa needed 11 more bytes: cap 1.11 -> 1.1102
with the measured 93,305 / 84,049 = 1.11013 and its reason beside it.
* docs: troubleshooting sections for the wave's new messages
Bun floor (setup, auto-update, /gstack-upgrade), /review stale base and
fingerprint temp-dir errors, the design image model override, the
unrecorded Chromium PID, and a contributor section for census report
lines (pass-rates flags, failure causes, headroom, cost unknown, overlay
v4, stale corpora, auq substance). A free test checks every
troubleshooting.md#anchor printed by setup or bin scripts exists.
* TODOS: reliability follow-ups deferrals; close the Bun floor, make-pdf $P and #2709 reap
* 0.4: 'A census went red' guide; docs for causes, session ledger, headroom and new messages
docs/evals/census-red.md walks one red line to a diagnosis: prerequisites, the
report fields, eval:pass-rates --run/--reds/--headroom with expected output,
what unknown/insufficient/unrecorded data means, free replay, and the single
'after a repair' paid run allowed by TESTING_INTERNALS. Linked from
CONTRIBUTING, TESTING_INTERNALS and both census reports.
TESTING_INTERNALS documents the failure_cause glossary (one row per
FAILURE_CAUSES value, diagnostic only under EVAL_POLICY v1), failure_detail,
sessions[], cost_known, the session ledger, the pass-rates views and the 85%
headroom alarm, main-only tracking issues, overlay contract v4, auq-matrix's
3-sample substance panel and the E5/E6 periodic cases. troubleshooting.md
gains #bun-too-old and an anchor per new user-facing message from lanes P, D
and F. README names GSTACK_DESIGN_IMAGE_MODEL. test/census-red-docs.test.ts
pins the glossary to FAILURE_CAUSES, parses every guide command and checks the
links.
* v1.91.20.0: CHANGELOG and version for the reliability follow-ups
* regen: SKILL.md and sections for all hosts, digest, context-budget fixture
bun run gen:skill-docs --host all; goldens and llms.txt re-rendered
unchanged; bun test/helpers/capture-context-budget.ts.
* integrate: register codex-multiblock-live and design-model-smoke (periodic, rule)
* 0.4: census report and eval:pass-rates views link the census-red guide
* test: cookie approval fixture reverses the Bun install-hint pin (E1)
The historical approved cookie-workflow prompt carried BUN_VERSION=1.3.10;
the regenerated install hint now pins 1.4.0, so the fixture's reversal
list gains that edit like the earlier waves' entries.
* test: release floor counts the two new periodic cases (E5, E6)
* fix: /freeze, /guard, /unfreeze and /investigate work on Codex and other env-var hosts
Templates quoted "$HOME/.claude/skills/gstack/...", which the env-var hosts'
rewrites turned into "$HOME/.agents/skills/gstack/..." on Codex (a path a
global install never has; same class on the other hosts), and setup's
runtime roots had no freeze/bin. Quoted runtime paths now use
{{RUNTIME_ROOT}}: $GSTACK_ROOT on env-var hosts (the fence gets the
runtime prelude), the unchanged $HOME path on Claude. _link_runtime_dists
links freeze/bin and the careful/bin helper it sources into every
env-var host's root. Also fixes the same guess in /benchmark-models,
/retro and /diagram. test/env-host-runtime-assets.test.ts fails on the
previous code: no fence guesses a $HOME/<dir>/skills/gstack path, every
$GSTACK_ROOT path a fence uses is in the host's runtime root, and the
installed Codex /freeze and /unfreeze fences run from a fresh shell.
Template-reading tests now read the Claude render or expect the
placeholder (deliberate change).
* CHANGELOG/TODOS: lane W's workflow, cases and guide; the env-var host freeze fix
* E6: a block that reads a SKILL.md is not a missing install
The first live codex-multiblock-live run passed its real check (the /learn
block found the install and printed the seeded learning) but the oracle
flagged a sed of the learn SKILL.md, whose prelude source quotes the
'gstack: no install found (tried $_r)' text. Only an emitted line that
starts with the message and names a real root counts now; free replay
negative added.
* test: qa-only fixture keeps its caller-set 30 s deadline
The qa-only-no-fix fixture tells the model to start the guard with 30 s
(a caller limit), so its deadline assertion's default stays 30 000 ms; the
product's Browser Quick default (180 s) does not apply to it. Reverts the
helper and its test to their base.
* C2: keep PROBE_DIR 'owned' (qa-functional-prompt pins it); qa measured 93,311
* test: paid file count and periodic floor include the two new periodic cases
* test: ship docs freshness pin checks meaning after A1's wording trim
The snapshot still hashes selected release paths (--select narrows them),
generated outputs and docs; the pin now checks that instead of the
pre-trim sentence.
* E3: reap the recorded Chromium only after the daemon has exited
With the PID now recorded (E3), browse stop reaped Chromium while the
daemon was still shutting down; the daemon read that as a crash, exited 1
and left its state file (chromium-profile-isolation's headless stop test
failed 3/3 locally, passes on base). The CLI waits up to 5 s for the daemon
to exit before reaping a survivor.
* regen: qa and qa-only exploratory sections (PROBE_DIR wording)
* B1: batching counter reads a target declared mid-line; replay corpus
GSTA-23's census 37228573062 failed plan-eng-multi-finding-batching with
reviewCount=0 after 975 s although D1-D10 were asked one per finding,
each with its saved decision-ledger record. The report wrote 'Report
destination requested by the user. Review target (fixed): `PLAN.md` ...'
and the counter read target fields only at a line start, so no brief
inherited the plan identity. A target field may now open any sentence of
its line; a mid-sentence mention or another file still gives no identity.
Floor (3), budget and kind unchanged. Corpus: 9 earlier passing censuses,
the red (now credited at D5), and a derived batching refusal; 975 s noted
in the headroom diagnosis.
* B1: shared-libs corpus labels its git path and covers the supported shape
Every census entry was recorded in CI's eval image (git 2.43), where
gstack-safe-git falls back because --no-lazy-fetch is rejected; each entry
now says so (git_path: fallback). Two synthetic supported-path entries
(HEAD and a history citation in the brief, as git >= 2.44 produces) keep
the 37193478719 pass and the 37195203538 refusal, and a test asserts both
shapes are covered and the actor reads no git-path-specific text.
* v1.91.21.0: move off v1.91.20.0 (claimed by #3031)
* test: settle nav-guard commands before asserting; make two new tests Windows-safe
navigation-guard: Bun's expect(promise).rejects slows each Playwright round-trip
to ~1s while it waits (50 page.evaluate: 43ms plain vs 46.6s inside the matcher),
so the click's 5s timeout expired before the guard saw the link-local request
(CI: 'click: Timeout 5000ms exceeded ... waiting for scheduled navigations').
Settle the command first, then assert on the refusal; a source check keeps it so.
env-host-runtime-assets: PROBE_FILES now matches either path separator.
detector-corpus ship-docsync-late-result: JSON-escape the <ROOT> substitution so a
Windows temp path's backslashes do not break JSON.parse.
* fix: CI run/job ids are not phone numbers; context-recovery fixture gets the full runtime bin
pii.phone.e164 redacted the run id in the published eval report's fetch command
(--run 37235771700), breaking the command it prints. Digit-only spans in a run/job
flag, gh run argument, actions/runs/ or /job/ URL segment, or run_id/job_id key are
now exempt; the same digits elsewhere still report.
context-recovery-artifacts: the fixture's ./bin/ lacked gstack-skill-start and
gstack-context-recovery (the preamble's runtime scripts since v1.71), so the agent
spent its 10 turns hunting for them and hit error_max_turns with no final message
(run 37235771700, 11 turns). Link the whole runtime bin; diagnostic rerun passed
in 6 turns.
* test: replay run 37237194905's late-result capture through the docs authority detector
The only failure was call 20's write to a mistyped shard root (3GanyY for 3GanyU),
which the model disclosed; the detector is correct and the A1 post-snapshot
--out under the fixture home stays permitted. Named red recorded in TODOS.
* test: keep docsync-authority in the Windows profile
The replay test's literal helper path tripped the Windows curator's shebang-spawn
pattern, dropping the whole file from the Windows profile (test-free-shards
invariant). The snapshot --out case is already covered by docsync-command-grammar,
so it is removed here, and the Linux-path replay skips on win32.
* fix: gstack-docs-candidate --help works; docs gate red lines name the denying tool failure
PR run 37239016858 ship-docsync-completion: the parent ran
'gstack-docs-candidate --help' (the helper's usage error points at --help, which
failed). That undeclared command made docsToolFailures non-empty, which denied
the atomic-replacement proof of the child's legitimate native Edit, so the red
line blamed the Edit's temp file. A replay of the capture proves the atomic
replace once --help is declared.
- the helper prints its usage for --help/-h and exits 0
- the fixture interface and command grammar admit the helper's bare --help
- the atomic-proof denial names its check and the tool failures behind it
(deny labels renumbered to their lines)
- replay with negative controls (undeclared command, temp left behind, renamed
elsewhere); diagnostic rerun of ship-docsync-completion passed
* test: retro-base-branch fixture uses main and a shallow report path
PR run 37241493377: the agent wrote retro-output.md one directory above the
requested <tmp>/retro-base/retro-output.md (it dropped the retro-base segment),
so the report check found nothing. The fixture also ran plain git init, which
makes master on CI's git while the prompt says the repo is on main; the agent
logged that mismatch as a learning. Init on main and use the temp dir itself as
the repo. Diagnostic rerun passed (18 turns, 203s).
* test: qa-quick sets a caller probe limit and drains past its session budget
C2 raised Browser Quick's native budget to 180 s. qa-quick is the only gate case
on the native budget, and its wall went 194 s (main periodic, pre-C2) -> 254 s and
274 s (this PR's runs) -> 300 s session timeout (run 37242691913): the agent now
probes every navigation target with a checkpoint, then needs ~80 s to report.
Like qa-only-no-fix, the eval now sets a caller probe limit (30 s), scopes the
fixture, and gives bun's outer timeout the session drain margin instead of racing
the 300 s session. The product's 180 s budget and the session budget are unchanged.
Diagnostic rerun passed in 195 s.
* review/ship: use main's post-expiry smoke sentence; v1.91.24.0
GSTA-23's measurement on the review workflow judge: main's 'Post-expiry smoke
rechecks are not-run.' passed 10/10; the C2 wording passed 6/10 and the
'..., even in 4c.' variant failed in PR run 37248650879. qa-probe-gates pins
main's sentence before step 4c.
* test: record the plan-design corpus test in the test-of-test baseline
#3031's ratchet now counts tests that slice a paid test file's source; the B1
plan-design-review-plan-mode replay corpus reads test/skill-e2e-design.test.ts
like its sibling corpora already listed.
* fix: --case no longer deadlocks importing the running CLI
runCaseDiagnosis loaded runPaidShards with a dynamic import of
scripts/test-paid-shards.ts; when that file is the entry it is still in its
top-level await, so the import never resolved and `--case` hung after its
first line (on main since #3031, found by the per-case measurement runs). The
CLI now passes its runner in; library callers keep the lazy import.
* test: name the in-flight file and the native error when bun dies before its summary (#3037)
* Bun 1.4.2 everywhere CI and setup pin the tested version
Bun 1.4.0 on Windows aborts about 3% of free-suite shards with native
bad-handle errors (main too; #3037); the same head on 1.4.2 had none in 180
shard runs. Every current 1.4.0 pin moves together, as bun-version-drift
requires: setup-bun in every workflow (CSO orchestration included), the eval
image (sha256 from bun-v1.4.2's SHASUMS256.txt, re-hashed locally), GitLab,
the Ubicloud free-suite setup, gstack-bun-version.sh's tested version,
engines.bun and @types/bun, the install hint in generated skills, the Dia and
cookie native-qualification pins, and docs and tests that state the current
pin. Recorded 1.4.0 observations, fixtures, CHANGELOG history and the CSO
images' oven/bun:1.3.10 base stay.
* fix: --case runs a single-case file whole
A session option naming the case (testName: 'office-hours-section-loading')
made caseSelection pick test-name mode, whose filter matched no Bun test, so
every trial would run hollow. A file that registers only this case now runs
whole before any name match is tried.
* fix(office-hours): reviewer returns a hash-bound receipt instead of echoing its verdict
The spec reviewer now Writes its verdict once, seals it with
gstack-office-hours-review seal (validates the saved JSON and prints
OFFICE_HOURS_VERDICT round=N sha256=<hash> path=<verdict>), and returns
only that line. check --receipt recomputes the hash and fails closed on a
missing, malformed, stale-round, wrong-path or mismatched receipt. The
saved file stays the sole findings inventory; validators bind the receipt
to the saved bytes and the judge reads those bytes.
* fix(office-hours): only blocking findings send a design back for another review round
Review findings carry severity blocking|minor (verdict schema v2). PASS when
no blocking findings remain; minor findings are recorded in Reviewer
Concerns and the Spec Review table, never a reason for another round.
CONVERGENCE needs a persisting blocking prior. A persisting or unverified
blocking prior cannot be relabeled minor. The 3-round cap is unchanged.
* fix(office-hours): rounds 2 and 3 are delta re-reviews of the captured design diff
prepare saves each reviewed design as round-N.design.md and gives later
reviewers the exact line diff since the previous round. A new later-round
blocking finding must cite (changed_text) an excerpt of a changed line;
seal, check and the validators verify it against the snapshots and fail
closed when a citation or snapshot is missing. Other new observations are
minor. Severity, anti-downgrade and CONVERGENCE rules are unchanged.
* test(office-hours): preservation judge accepts COMPLETED with recorded minor findings
* fix(office-hours): parent fixes only blocking findings; callers may set --max-rounds
Between rounds the parent edits for blocking findings only; minor findings
stay recorded. gstack-office-hours-review takes --max-rounds 1|2|3 (default
3) on every command; reaching it stops at MAX_ITERATIONS. The marathon
fixture sets 2 rounds so round 1 (full) and round 2 (delta) run live.
* docs(changelog): office-hours review loop changes in v1.91.28.0
* fix(cso): helper JSON keeps run URLs only in the validated workflow slot
The shared redactor now treats GitHub Actions run ids as CI metadata
(c245d1bc2), so main's #3041 invariant that the same run digits stay
redacted outside qualification.workflow held only through the phone
false positive. Make it explicit in the CSO sanitizer.
* test(plan-ceo): split-overflow progress detector accepts the skill's candidate wording
Collection never ended when the model wrote 'E1 Slack:' / 'E1-SLACK:' or
labels like 'Include in this quarter' / 'Defer with flip trigger', so the
session ran the whole review (930-1464 s) and failed on its extra findings
or the case deadline (6 of 21 runs). The candidate check now accepts a
platform-bound E-id and the action is the bucket's leading verb, as the
skill names them (Include/Defer/Cut/Hold). The semantic judge, call floor
and include/defer/cut validation are unchanged. Replays the captured CI
37374526437, t2, t9 and t15 menus with negative controls.
* test(plan-ceo): split-overflow candidate identity is the one E1-E5 token
The skill fixes no header or question wording, and shape-matching kept
missing new forms (header 'D1.1 Slack' with 'SCOPE-E1:', 'D1.1 E1 Slack',
'E1 Slack DM bot ...:', a misspelled 'Mattermst'). Identity is now exactly
one E1-E5 token across the header and the question's first line, with no
other candidate's platform named. Options rules, judge, call floor and
include/defer/cut validation are unchanged. Replays all 31 measured runs;
the quoted-row controls now quote a second ledger id.
* v1.91.30.0: move past #3044's claimed 1.91.29.0
* test: register the split-overflow wording replay as a free fixture
* fix(skills): resolve judge-cited ambiguities in /qa and /plan-eng-review
- tasks JSONL keeps the raw branch as GIT_BRANCH instead of overwriting
gstack-slug's sanitized BRANCH used for artifact names
- the Codex outside-voice block runs for ready or unverified, matching
its routing rule
- plan-eng startup names the preamble's own setup prompts; the
cross-project question waits for Step 0
- /qa 8c: a fix that applies but leaves the finding reproducing gets one
revised fix, else it is reverted; browser verification names its before
screenshot; the exploratory loop drops its duplicate 3-step summary
* fix(plan-eng): keep the startup steps in their pinned order
* fix(plan-eng): explicit spawned sessions follow the spawned rule at the scope gate; one announcement rule
A missing tool still sends the prose menu whatever the session type, but a
session the caller or host explicitly marks spawned selects the menu's
Recommendation without prose or BLOCKED. The pre-load announcement
sentence now states Step 0's single check.
* eval policy v3: judge panels gate on the per-dimension median of 3 samples
A judge dimension now passes when at least 2 of its 3 samples meet the
unchanged threshold (judgePanelMedian); any other sample count or a
non-numeric sample fails closed. Thresholds, sample count, judge model and
prompts are unchanged; panel logs show every sample and the mean, and name
the median as the gate. Applies to the skill-llm-eval panels, the
workflow-judge cache, the auq-matrix substance panel and the calibration
replicas. EVAL_POLICY 2 -> 3 and HARNESS_VERSION 1 -> 2 (verdicts change),
so every case starts a new pass-rate series. Replays the captured panels
from jobs 112032744416 and 112049856586 as replayed evidence.
* test: delete test-infrastructure dead code (G)
- exit-propagation drives the runner's real strict verdict
(BunTestOutputClassifier + strictTestExitCode); delete the unused
shardRunLooksTruncated predicate.
- delete skill-coverage-matrix registry + its gate (nothing reads it; the
floor already iterates skillCensus()).
- delete touchfiles-facade export-parity tests (Bun fails missing imports
at link time) and the duplicated E2E_TIERS tier-value test.
- delete brain-cache-spec TRANSPORT_DEFAULT_POLICY, SKILL_RUN_RETENTION_DAYS
and the now-unused BrainTrustPolicy type with their literal tests.
AUTOPLAN_PREFLIGHT_BUDGET_BYTES stays: skill-preflight-budget enforces it
against real resolver output.
- delete audit-compliance's JSDoc-comment grep.
* test: replace product tests that fake the product with real-boundary tests (F)
- design: serve.test.ts drove an inline mirror server; now two tests run the
real serve() on an ephemeral port (reload confinement, submit exit 0).
- setup-gbrain: rollback + voyage tests execute the template-extracted init
blocks (3 sites) instead of drifted local bash copies.
- terminal-agent: internalHandler source greps replaced by a behavioral
/internal/grant + /internal/revoke auth matrix (no/wrong/valid token).
- /health: server-security-surface and the server-auth / security-audit-r2 /
sidebar-tabs source greps fold into one liveness-only check on the real
body; the L4 sidecar wiring gets a behavioral /pty-inject-scan test.
- delete tautologies (browser-manager onDisconnect, memory-command #12),
ios swiftui tap fixture self-check, memory-ingest put_page grep, detach
source greps, sidebar-agent absence pins, dead-CSS pins + the dead CSS,
security-audit-r2 Task 1 + the test-only meta-commands re-export,
duplicate generated-SKILL.md checks.
- make-pdf coverage-gaps cases move into their owner test files.
* test: delete tests of dead eval code (A)
- A1: the retired Eng lexical oracle (evaluateEngSeedCoverage,
isEngSeedDecisionAUQ), the completion-handoff detector and the retained
corpus had no paid caller since v1.87.6; delete their 26 replay files,
~2.6k helper LOC and fixtures, and the dead blocks in 8 mixed files
(live hasNativePlanTerminal / batching assertions stay).
- A2: dead viewport approvers in autoplan-artifact-permission and their 11
replay files + fixtures; recorder/launcher cases stay.
- A3: never-wired oracles and seeders (autoplan-phase-order,
eng-finding-fixture, ceo-paired-fixture, design-ui-scope,
plan-skill-completion, pty-current-screen, required-reads,
transcript-section-logger); plan-seed-submission now decodes through the
production createPtyScreen; section manifests name their actual guard.
- A4: zero-reference helper exports, plus execGit and invokeAndObserve
found by the reachability pass.
- 52 fixtures orphaned by the deletions; touchfile and selection-table
entries for every deleted path.
* test: clean up the paid eval lane (B1-B4, B6, B7)
- B1: delete paid files that assert nothing or cannot pass meaningfully:
skill-llm-eval-spec and skill-e2e-spec-execute (test.todo), gemini-e2e
(+ gemini-session-runner; no gemini CLI in CI), ship-idempotency (red
since v1.63), the two opus-4-7 *-sonnet overlay wrappers, conductor-prose
(+ its source-evaluation replay), codex-e2e-plan-format; drop their keys,
scripts and census rows.
- B2: skill-llm-eval grades browse/sections/command-list.md with one union
judge that also carries the baseline score pin; regression-vs-baseline
deleted (paid run: pass, c4/c4/a4).
- B3: memory-pipeline, ios-qa, ios-qa-swift-build and plan-tune-cathedral
make no model calls; renamed out of the paid glob so they run on every
PR. Swift builds need GSTACK_TEST_SWIFT=1; device stub deleted.
- B4: codex-e2e*, outside-voice, aside and ios-device cannot run in the CI
image; excluded from the weekly lane with a tracked re-entry condition.
- B6: fold opus-47's negative routing controls into skill-routing-e2e
journey-negatives (paid run: 3/3 unrouted) and delete the file.
- B7: delete the never-green brain-privacy-gate eval; a free
gstack-skill-start test now proves consent precedes artifacts egress.
* test: retire the finding-count cluster and trim its helpers (C)
- C0/C1: the five never-green evals (skill-e2e-autoplan-chain and
skill-e2e-plan-{ceo,eng,design,devex}-finding-count) failed on harness and
budget, never on skill behavior; delete them, their touchfile/tier ids,
AUTOPLAN_CHAIN_BUDGET and the dedicated eighth periodic slice (--slices 7).
- C2: delete the helper groups whose only paid consumers were those files
(11 modules), trim claude-pty-runner and eng-seeded-coverage to the paid
closure, and delete the free replay tests whose assertions exercised only
that dead code (89 files, 135 orphaned fixtures). Blocks that used dead code
only as input for a live subject keep their assertions: the multiSelect
default moved to plan-review-decisions, runner PTY tests use inline caller
policies, and the timer-safe budget checks moved to eng-finding-retry-budget.
- The eight production-touching files stay except ceo-current-decision-record
(its template read only feeds the retired counter).
- CARVE_GUARDS.autoplan is behavioral 'none'; TODOS records the lost chain
and per-finding cadence coverage with their re-entry tests.
* test: fold per-incident replay series into their detector owners (D)
Twelve detector families move into one owner test each: 73 incident files
become describe blocks in ceo-section-loading-fixture (stale-fill race),
model-overlays, coverage-audit-evidence, autoplan-phase-observer,
native-auto-decide, outside-voice-evidence, eng-first-review,
plan-count-completion, plan-count-file-permission, ceo-mode-option,
plan-scope-selection and plan-count-prerequisite. Each block keeps its original
code and fixture, so every case still runs; only tests asserting the incident
file's own touchfile registration are dropped (41). Touchfile lists that named
an incident now name its owner.
* test: start the plan-count history PTY on its readiness marker (H)
The fake CLI prints a startup marker and the runner waits for it instead of the
fixed 8 s startup sleep (8.6 s -> 0.9 s locally). eng-semantic-terminal's
sleeping registration cases went with C; plan-count-timeout keeps the fixed wait
because it asserts deadline behavior.
* test: derive paid touchfiles from each eval's static closure (E)
touchfiles.test.ts now checks, per key, that the paid file's static
test/helpers and test/fixtures closure (plus fixture paths it names in string
literals) is covered, and names the file, path, chain and key to fix when it is
not. Free *.test.ts files are no longer touchfiles, so editing a free replay
test stops selecting paid evals: 950 entries removed, 653 real closure paths
added. The hand-copied inventories go: periodic-fixture-selection,
fake-impeccable-touchfiles and 45 per-file selection examples. Selection for
the sample edits (plan-eng-review template, claude-pty-runner,
plan-count-fixture, gstack-config) loses no case under either profile.
CONTRIBUTING documents the rule and its lower bound.
* test: skip hollow tier shards and census judges in the paid planner (B5)
A paid file is now skipped for a tier lane only when every E2E id it registers
is known statically and none has that tier; ids come from the touchfile
registrations and literal testName/*IfSelected arguments, so a comment or
skill path that quotes another id cannot unschedule it, and computed names
keep today's scheduling. --list and the manifest show each skip as
"skipped: no E2E_TIERS id has tier <tier>". The weekly gate census drops the
LLM judges (--skip-judges); they still run in the periodic census and PR gate
lanes. Gate lane 52 -> 42 files, census 41; periodic 77 -> 69.
* test: run seven paid evals on the current default capture model (B8)
skill-e2e-{auq-matrix,plan-format,qa-bugs,retro,workflow} pinned
claude-opus-4-7 and skill-e2e-office-hours plus -brain-writeback pinned
claude-sonnet-4-6; none tests a historical model, so they now capture with
resolveEvalModel('capture'), and the free harness tests that execute these
registrations receive the same resolver. The paid re-pin run passed all of
them. skill-e2e-{design,office-hours-phase4,plan-prosons,plan} keep
claude-opus-4-7: six of their cases failed on the default model (three
timeouts, a missing report file, a format miss and a posture score of 3), so
per the plan's fallback they keep their pins with a TODOS entry. The pre-spend
estimate and drop threshold are in docs/test-audit-2026-09.md.
* test: guard the reduced suite against new test-of-test files
- test/test-of-test-ratchet.test.ts records the 228 free tests that import only
test/ code and fails on a new one, naming the owner test to extend instead;
a stale baseline entry fails with the remove instruction.
- test/helpers/resolve-repo-path.ts is the one specifier/literal resolver for
the ratchet and the touchfile closure invariant, with its own unit tests.
- CONTRIBUTING "Test tiers" describes the paid-failure workflow (fix, then one
row in the detector's owner test) and the ratchet; TEST_PORTFOLIO gains the
detector -> owner-test table and no longer claims an Autoplan chain eval.
- TODOS: automatic exclusion policy for chronically red periodic files (P3),
the deferred native-completion table collapse, the unused CEO payment
seeder; the PTY readiness item is narrowed to the paid runner.
- docs/test-audit-2026-09.md collects the triage, security mapping, inventories,
selection proof, behavior-commit decisions and retained false positives.
* v1.91.8.0 test: smaller suite, derived paid selection, retired never-green evals
Release metadata for the test-reduction branch: VERSION 1.91.8.0 (1.91.7.0 is
claimed by #2983), CHANGELOG with the measured before/after table and a
contributor section, durations re-recorded on Ubicloud standard-16 (857 files,
0 failures), the agents digest, CONTRIBUTING's after-measurement row, the B8
fallback TODOS entry, and the after metrics, kept-vs-plan notes, B8 run and
census estimate in docs/test-audit-2026-09.md.
* fix(ubicloud): skip retrieval globs that match nothing instead of reporting a failed pull
* test: count issue-numbered Design findings in the UI-scope gate eval
* fix(ci): free-tests lane actually runs the make-pdf e2e gates
The 9 make-pdf/test/e2e gate tests probe make-pdf/dist/pdf,
browse/dist/browse, and the diagram-render bundle, then self-skip when
absent. The required free-tests lane never built any of them, so the
gates silently skipped on Linux for their entire life (verified: 9 of
14 skip, exit 0). make-pdf-gate.yml's justification for deleting its
Linux leg claimed the free lane covered this — it didn't.
- new build:gates script: exactly the three artifacts the gates probe
(full bun run build compiles five binaries; ~60-90s tax on the only
required check is not warranted)
- free-tests.yml: build:gates step + poppler-utils +
fonts-noto-color-emoji (fonts must precede the first browse daemon
launch — Chromium snapshots fontconfig at startup; verified live:
a warm daemon renders tofu, a fresh one embeds NotoColorEmoji)
- make-pdf/test/e2e/ci-prereqs.test.ts: GSTACK_EXPECT_BINARIES=1 (set
by the workflow) inverts the skip polarity in CI — dropping the
build step or poppler fails the lane instead of re-opening the
silent-skip hole
Pre-flight: all 9 gates green on Linux locally.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): kill the three zero-test eval jobs (hollow green)
- delete the vestigial e2e-codex / e2e-gemini matrix rows: both files
are whole-file periodic-tier, so with no row tier: they ran ZERO
tests and reported green on every PR (~2 min of runner each, pure
false confidence; the periodic lane owns those suites)
- e2e-pty-plan-smoke gains tier: gate — its two files are whole-file
describeE2ETier('gate'), so the job burned ~7 min of container setup
then skipped every describe
- KNOWN_TIER_UNSET burned down to empty; the ratchet stays armed so a
future row/file tier mismatch fails the suite instead of shipping
hollow green
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): least-privilege permissions + fork-safe concurrency keys
- evals.yml / evals-periodic.yml evals jobs: explicit contents:read +
packages:read (container-image pull) and persist-credentials:false —
the jobs that execute PR-authored code with three provider API keys
ran on the repo-default token grant with the token written into
.git/config
- permissions blocks for the 4 workflows that had none (skill-docs,
make-pdf-gate, windows-free-tests, windows-setup-e2e)
- fork-safe concurrency keys: actionlint, skill-docs, make-pdf-gate,
windows-setup-e2e switch from head_ref to PR-number keying — a bare
branch name carries no fork prefix, so same-name branches from two
forks shared one group and cancelled each other's runs
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): one bun version everywhere + drift tripwire
Lanes disagreed four ways: 1.3.13 (free-tests, windows, Dockerfile.ci),
latest (quality-gate, make-pdf-gate), unpinned (skill-docs,
version-gate — setup-bun installs latest), 1.3.10 (.gitlab-ci.yml).
Different Bun versions change the runner output shapes the strict
classifiers regex-match, spawn semantics, and shell parsing — a lane on
a different Bun tests a different product; Dockerfile.ci's own comment
records this class biting once already (silent 1.3.13/1.3.14 drift).
All surfaces pinned to 1.3.13; test/bun-version-drift.test.ts scans
every workflow setup-bun stanza + Dockerfile.ci + .gitlab-ci.yml and
fails on any mismatch or unpinned stanza. skill-docs also gains
--frozen-lockfile (was bare bun install).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(ci): bind the three-way image-tag hashFiles() expressions
evals.yml, evals-periodic.yml, and ci-image.yml each compute the CI
image tag from hashFiles('.github/docker/Dockerfile.ci', 'bun.lock',
'patches/**') — synced by comment only (TODOS.md 'CI three-way
image-tag drift'). If one input list drifts, that workflow computes a
different tag for the same content: eval lanes silently rebuild the
image every run, or ci-image prebuilds a tag nobody looks up. The test
extracts each tag-computation site and fails on any mismatch.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): ci-image stops rebuilding the identical image every ship
- package.json out of the trigger paths: the tag hash deliberately
excludes it (version bumps every ship), so every merge rebuilt and
re-pushed the IDENTICAL tag (~2m26s for zero content change);
patches/** added (it IS a tag input)
- manifest existence check (mirrors evals.yml): tag already exists →
skip the build
- concurrency group: two rapid main pushes raced pushing the same
:latest/:buildcache tags
- cron staggered 06:00→04:00 Monday: it shared the exact minute with
evals-periodic, which could race a half-pushed tag or duplicate the
build
- timeout-minutes: 30 (was unbounded → 360-min default for a hung
docker build)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): quality-gate drops the 74s full-history checkout
fetch-depth:0 cost 74 of the job's 92 seconds; the three gates it feeds
take ~12s combined. Shallow checkout + exact-SHA fetches for the diff's
base/head (an exact-SHA fetch, not a guessed depth — long-lived
branches and merge queues still resolve), with a --deepen fallback for
push events whose 'before' is unusable. timeout right-sized 20→10 min.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): small-lane batch — timeouts, right-sizing, windows cache warm-start
- timeout-minutes on the 6 remaining unbounded jobs (actionlint 5,
skill-docs 10, version-gate 10, make-pdf-gate 15, pr-title-sync 5,
evals build-image 15) — a hung step sat on GitHub's 360-min default
- right-size measured-over-long timeouts: dependency-review 10→5,
windows-setup-e2e 15→10
- dependency-review: 2-core runner (28s API call on an 8-core box) and
drop .github/workflows/** from its trigger paths (workflow edits have
no dependencies to review)
- windows caches gain restore-keys: a lockfile bump paid the 26s/43s
restore for a guaranteed cold miss
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): scope GSTACK_HOME to each file's execution window
Five files assigned process.env.GSTACK_HOME at module scope. Shard
processes evaluate sibling modules before running their tests, so the
assignment leaked into every other file in the shard — the damage was
already visible in defensive workarounds (relink.test.ts:28 'fresh
install test saw a neighbor's skill_prefix'; cdp-e2e's own comment
documents a sibling's temp dir baked into artifacts).
Pattern: save original, assign in beforeAll, restore in afterAll
(cdp-e2e already restored but still assigned at load — its window now
matches the others). GSTACK_TELEMETRY_OFF and GSTACK_PROJECT_SLUG get
the same treatment where they rode along. Victim files' defenses stay
in place (cheap insurance).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: tripwire against module-scope GSTACK_HOME assignments
Column-0 assignment of GSTACK_HOME / GSTACK_STATE_ROOT in any tracked
*.test.ts fails with the file:line and the fix (beforeAll + afterAll
restore). Kills the cross-file env-leak class the previous commit
swept.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): e2e-harness-audit derives its skill census from disk
The hand-maintained 39-name SKILL_GLOBS list had drifted to 39 of 54
SKILL.md.tmpl on disk. No live gap today (none of the 15 unlisted
skills is interactive), but the next interactive skill would have
landed unguarded with zero signal. The audit now walks top-level dirs
for SKILL.md.tmpl (statSync so symlinked dirs like connect-chrome
count), so new skills are in scope the commit they appear.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(evals): judges honor the eval-model resolution chain + real 429 backoff
callJudge inlined GSTACK_EVAL_MODEL_JUDGE || sonnet, silently ignoring
the global GSTACK_EVAL_MODEL override every other eval call site honors
via lib/eval-model.ts. New 'judge' kind in DEFAULTS (sonnet — the D1a
pin-on-regressors calibration stands; model CHOICE unchanged) and
callJudge resolves through it: explicit arg > GSTACK_EVAL_MODEL_JUDGE >
GSTACK_EVAL_MODEL > default.
429 handling upgraded from one fixed 1s retry (reliably lost races at
CI concurrency) to three jittered exponential retries (~1s/4s/16s),
honoring the server's retry-after when present.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): the two expect(true) paid stubs become test.todo
skill-e2e-spec-execute (600s budget) and skill-llm-eval-spec (300s)
reported PASS on every periodic run while asserting nothing. Deleting
them would remove the periodic-tier selector surface they exist to
register (diff-based selection for spec/ changes), so they become
test.todo — reported as todo/skip, never pass — with the v1.1
implementation specs kept in-file.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): reactivate 5 quarantined browse tests (2 security)
extension-sender-auth's two privileged-message denial tests (content
script + missing sender.url — the extension's security boundary) and
snapshot's three skips were quarantined 'pre-existing' failures. Root
cause: machine-local state on the quarantining dev machines — the test
and gate code are byte-identical between the quarantining commit
(410b4928) and HEAD, and all five pass deterministically on a clean
checkout (68/68 across both files, multiple runs). No assertions
weakened, no product changes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(evals): activate the 4 paid test files that could never run anywhere
carve-section-loading, codex-e2e-plan-format,
codex-e2e-recommendation-substance, and llm-judge-recommendation gated
on EVALS/tier (free suite loads them as describe.skip) but their names
fell outside PAID_TEST_GLOBS, so no paid lane ever selected them — net
execution zero, forever. The existing matrix tripwire filtered on
isPaidTestFile() first, so it was blind to exactly this class (the same
bug that hid the pre-split monolith's gate tests for ~8 releases).
- PAID_TEST_GLOBS: codex-e2e* + skill-llm-eval* wildcards (replacing
exact names) + llm-judge-recommendation + carve-section-loading;
package.json's six test-script glob lists mirrored
- codex-e2e-plan-format gains the explicit periodic tier gate its
siblings carry (external-service rule) — without it the sharded
runner's no-guard default would spawn Codex in the gate tier per PR
- eval:bg:periodic --timeout 32400→37800: the census growth pushed the
periodic worst case to 35910s; the old value had 270s of headroom
BEFORE this change and would now kill healthy runs mid-flight
- new test/paid-orphan-tripwire.test.ts: any EVALS/tier-gated test file
outside the globs fails the free suite (reasoned SCANNER_EXEMPT for
the gate helpers + meta-tests) — the class-killer
- paid-shards pins updated: the four orphans now assert INSIDE the
census
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(browse): restrictDirectoryPermissions warns and skips symlinked dirs
Closes the Windows Free Tests red: recent lane failures showed a
platform-unguarded POSIX mode-bit assertion ('Expected: 493' — a
symlink-skip test) from PR-branch variants; the KNOWN_WINDOWS_SAFE
force-include reason ('mode-bitmask hits are POSIX-branch only') did
not hold for that shape, and main had neither the guard nor the
behavior.
- product: lstat first; a symlinked dir gets a warning and a skip on
both platforms — chmod AND icacls dereference the link, so
restricting through a symlink hardens an unvetted target (and
/inheritance:r could lock out its real owner). All callers already
treat hardening as best-effort (try/catch).
- test: the symlink regression test, platform-aware — symlinkSync in
the house try/catch skip pattern (Windows runners without Developer
Mode can't create symlinks), mode-bit assertion guarded off win32,
behavior assertions (no throw, warning text, target readable)
everywhere; POSIX still proves the skip (0o755 unchanged, not 0o700)
- KNOWN_WINDOWS_SAFE reason updated to the now-true premise
20/20 pass on Linux.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): unique tmp dirs for plan artifacts + audited live-repo cwd sites
Six paid PTY tests wrote their expected plan artifact to a FIXED shared
/tmp path ('/tmp/gstack-test-plan-<mode>.md') and rmSync'd it in
finally — under --retry 1, EVALS_JOBS>1, or two concurrent worktrees, a
sibling's cleanup deletes this run's artifact and the D19 'agent did
not produce expected plan file' assertion fires spuriously. Each test
now mkdtemps its own dir, interpolates the unique path into the agent
prompt (fixture-sourced prompts get a replaceAll + drift guard that
throws if the fixture's literal ever moves), and cleans up its own dir.
The 18 cwd:-into-the-live-repo sites were audited: all deliberate
(skill registry + hermetic pre-trusted dir, in-repo gen renders, git
history reads, slug resolution) — each now carries a
'// LIVE-REPO CWD: <reason>' comment so the next audit can tell
deliberate from accidental.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): trim the seven over-wall 1700s timeouts to the 1500s physical ceiling
1,700,000ms (28.3 min) exceeded every wall these tests run inside: the
25-min CI job timeout and the 1800s sharded-runner wall (which also
leaves --retry 1 zero room for a second attempt). Budget above the wall
is fiction, not headroom — a test that actually used it produced a
job-level kill (no bun summary, no artifact) instead of a clean
per-test timeout. No recorded p95 exists for this family (they are
being retiered to periodic in the re-platform wave); the trim stops at
the physical ceiling rather than guessing lower. Final policy lands in
the Wave-2 eval-budgets constants module.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(gen): main() guard — importing gen-skill-docs no longer regenerates the tree
The generator's whole body executed at module load, so any import of it
(test/gen-skill-docs.test.ts pulls assertSinglePreamble via require();
test/catalog-trim.test.ts imports helpers) regenerated all 71 SKILL.md
in place — the root cause of half the TREE_MUTATING serial-shard
entries (hazard class #2532). The body now lives in an exported
main(): number behind if (import.meta.main).
Semantics preserved exactly: failure exits are immediate (matching the
old top-level process.exit), success leaves the event loop to drain so
the llms.txt fire-and-forget IIFE finishes its write, and the module
stays synchronous/require()-able. Proofs: byte-identical --host all
output (git status clean), --dry-run stale-tree still exits 1 (the
skill-docs freshness lane depends on it), and the new
test/gen-skill-docs-import-purity.test.ts pins load-time purity via a
subprocess probe (mtime-based, so a dirty worktree can't false-fail).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(gen): --out-dir renders every host, outputs-only
--out-dir was Claude-host-only (gen-skill-docs.ts:842), which forced
the codex/factory-regenerating tests (gen-skill-docs, skill-validation,
host-config) to mutate the live tree — the reason they sit in the
TREE_MUTATING serial shard. The flag now mirrors ALL outputs into the
out-dir: external-host trees (.agents/.factory/... via
processExternalHost), external section files, openclaw docs, and
gstack/llms.txt (a catalog-mode render must never rewrite the tracked
index). OUTPUTS ONLY — inputs (templates, sections/, host configs) are
always read from ROOT, so an empty out-dir can never feed the render.
rewriteSectionBase stays Claude-only (external hosts have their own
path grammar).
Proofs: in-place --host all is byte-identical (tree clean);
--host all --out-dir <mkdtemp> renders the full multi-host tree with
ROOT untouched; gen-skill-docs-out-dir tests + 415/415
gen-skill-docs.test.ts green (bin/dev-setup's claude rendering
byte-compat).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(evals): every E2E key's dep list names its own declaring test file
129-of-177 keys omitted their own test file, so editing only a test's
prompt or assertions selected NOTHING — the changed test never ran on
the change that changed it. 135 keys self-registered (110 E2E + 25
LLM-judge), resolved by strict declaration evidence (testName:/
testIfSelected/judge call sites), with skill-name false positives
excluded.
e2e-tier-alignment's warn-only branch for unregistered files is now a
hard failure with a 4-entry KNOWN_UNREGISTERED ratchet (template-
literal testNames, fail-open-safe) + a burn-down test so the set only
shrinks. Selection sanity: a one-file diff on skill-e2e-qa-workflow now
selects its 4 tests (was 0); skill-llm-eval 0 → 25.
Known follow-ups (filed): 15 E2E + 2 judge PHANTOM keys select tests
that exist nowhere; codex-e2e-plan-format's testIfSelected names have
no map keys (run-all only).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): ratchet the 8 newly-visible gate-matrix gaps
The self-registration sweep made these eight files' gate-tier keys
visible to the census for the first time — their gate tests run in NO
CI lane today (pre-existing hole, newly measurable). Ratcheted into
KNOWN_MATRIX_GAPS with the burn-down note: the paid-lane re-platform
runs every gate file by construction and retires this ratchet class.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): duration-aware LPT shard packing for the free suite
Hash sharding balances file COUNTS (1.15x spread) but not cost — the
Playwright-launching files landed 4/3/4/1/2/1 across 6 shards, giving a
measured 28s–97s shard spread and ~40s of idle tail on every run.
Full-suite mode now packs by recorded per-file durations
(longest-processing-time-first) when the committed seed
scripts/free-test-durations.json exists.
- ONE store, no overlay: the seed is refreshed occasionally via the new
--record-durations mode (each file timed in its own child — exact,
and immune to bun's stream buffering, where silent passers print no
header to timestamp); GSTACK_FREE_TEST_DURATIONS overrides the path
for experiments; CI never records
- seed is a hint: missing → silent hash-shard fallback; corrupt (bad
merge) → one warning + fallback; unknown files → 75th-percentile
pessimism so a surprise long-runner can't recreate the tail
- packed shards get duration-aware walls (max(base, predicted x 3)) —
LPT decouples count from cost BY DESIGN, so the 5s/file heuristic
would undersize a shard holding few expensive files
- one log line per shard (files + predicted seconds) so packing
regressions are diagnosable from any run log
- the --shard CI-matrix path is untouched: stable hash indices are its
contract
- successor note in-code: bun >=1.3.14 ships native --timings/--shard
LPT — swap this packer when the repo unpins 1.3.13
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): decouple slop:diff from bun run test; quality-gate runs it per PR
'bun run test' silently appended up to two 120s npx slop-scan runs plus
a git worktree add/remove after the suite (2>/dev/null || true) —
invisible in the documented '~90-100s' timing and pure friction in the
pre-commit loop. Decoupling is not coverage removal: quality-gate.yml
now runs slop:diff on every PR (advisory, matching its in-repo 'never
blocking' contract), and /review already invokes it explicitly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): eval-budgets timeout tiers + fit/ceiling policy test
Five named tiers (JUDGE 120s / CAPTURE 300s / CAPTURE_LONG 600s /
PTY 900s / PTY_LONG 1200s) replace hand-ratcheted sprawl (46x300s,
46x120s, 44x360s, 44x180s, 27x240s, 19x150s, 13x420s, 12x600s...),
much of it inflated to paper over the old 40-way in-shard concurrency
that the sharded runner's 1-file-per-shard model kills. Policy test
pins: every tier fits the shard wall minus 120s overhead (the
structural fix for budgets-above-the-wall fiction), tiers stay ordered,
and no paid literal exceeds PTY_LONG x1.25 — oversized tests get split,
not budgeted past the wall.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): shared runBin helper for bin-script unit tests
~36 free test files each carry a near-identical local run() (spawnSync
+ utf-8 + {status, stdout, stderr}) differing only in env composition,
cwd, and timeout. runBin absorbs the invariant core; options carry the
variance (gstackHome sets BOTH GSTACK_HOME and GSTACK_STATE_DIR — the
config-precedence trap several locals rediscovered independently; home
for $HOME-anchored bins; input/trim/timeout/maxBuffer). Free-test-only
by design so it never becomes a de facto global touchfile. Migration of
the 36 call sites lands separately (mechanical batches).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): runBin trim assertion — trim shapes stream ends, not interior
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(test): mechanical sweep — 298 paid-test timeouts onto eval-budget tiers
69 files, both shapes (trailing bun-test budgets and runner
timeout/timeoutMs options), ROUND-UP ONLY so nothing that passed can
start failing: 75 → JUDGE_MS, 137 → CAPTURE_MS, 74 → CAPTURE_LONG_MS,
9 → PTY_MS, 3 → PTY_LONG_MS. Raw >=60s literal count in the paid scope:
395 → 97, of which 51 are non-timeout noise (fixture dates, run IDs)
and 46 are enumerated justified holds (comment-carrying calibrated
budgets, poll-loop constants, utility spawn waits, and the seven
physical-ceiling 1_500_000 sites). The eval-budgets policy ratchet
keeps the residue from regrowing.
Known collapse: where an inner runner budget and its enclosing test
budget now share a tier, the old stagger is gone — an overrun surfaces
as a bun test timeout instead of a graceful runner timeout
(diagnosability trade, not a correctness one).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: coverage fill — 95 tests for six zero-coverage surfaces
- eval CLI family (eval-list/compare/summary + eval-select smoke): the
primary interface to eval results had no tests; isolation via a fake
gstack-slug under a mkdtemp HOME (the scripts' real resolution path —
they do NOT honor GSTACK_EVAL_DIR; only EvalCollector does). Pinned
current behavior: eval-list does NOT exclude _partial runs (documented
improvement candidate)
- slop-diff (runs on every /review + quality-gate): fixture git repo +
first-on-PATH npx stub (never downloads real slop-scan); no-diff
early exit, missing-scanner fallback, fingerprint line-insensitivity,
merge-base worktree scan
- bin/gstack-code-intelligence CLI arg surface (lib was covered, the
284-line CLI wasn't): select/consent/suggest/index/search gating;
pinned: --help routes to usage failure exit 1 (no handler)
- browse media-extract: the page.evaluate callback exercised in-process
against a mock DOM (no exports added) — lazy-src fallback chain,
HLS/DASH detection, bg-image url() parsing, 500-element cap
- browse session-cookie-store: factory contract (cookieName/ttlMs/
maxSessions eviction, cross-store isolation, mint→validate
round-trip); store is in-memory — no fs cases exist
- lib/version-source direct unit tests (gstack-version-bump.test.ts
spawns the bin, never imports the lib): parse/format/cmp/bump
coercion, npm 4→3 translation, #2501 mangled-JSON regression class
All hermetic (mkdtemp homes, runBin child isolation); windows curation
correctly partitions the six.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(test): first runBin migration batch (3 of ~36 run() duplicates)
explain-level-config, benchmark-cli, evidence move onto the shared
helper; each file's remaining special-case spawnSync sites (raw-buffer
probes, env-scrub probes) stay put deliberately. 55/55 green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(evals): paid shards spool to disk + shared runShardChild lifecycle
- runPaidShard no longer buffers whole 30-min stream-json streams in
RAM (x concurrent jobs): every byte tees to a per-shard log file
(slug-named, path printed at START for mid-run inspection and on the
FAILED terminal line); failures print a 64KiB tail read back from
disk; passing shards stay quiet (the file is the record) — the free
runner's proven contract. Classification unchanged: the strict
classifier still sees every byte first.
- the ~35 duplicated spawn/group-kill/wall-timer/finally-reap lines
move into runShardChild in test-strict-output.ts (detached-per-
platform spawn, signal forwarding, SIGKILL group kill at the wall,
drain-before-verdict); designed so the free runner can migrate later
- expectedFiles drift fixed toward ENFORCEMENT: the injected-command
exemption is gone — a fake command exiting 0 without bun's terminal
summary now reads FAILED (pinned: silent-pass → failed)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(evals): parent-computed selection propagates to shard children
The sharded runner computed diff selection once, then each of its 48-73
children recomputed it at module load — including, on touchfiles-diff
branches, a per-child bun subprocess evaluating the old data file (20s
timeout each). The parent now serializes {version, selected, reason} as
EVALS_SELECTION_JSON into the shard env; e2e-helpers adopts it at load.
Fail-open preserved: any parse/shape violation → ONE stderr warning +
local recompute; absent env → silent local compute (non-sharded
entrypoints unchanged). Drift test pins parent→child round-trip to
identical selection decisions plus the malformed/absent cases.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): kill the four worst fixed sleeps (300s/30s/30s/20s)
- watchdog.test: the 20s blind wait for one production parent-watchdog
tick becomes BROWSE_PARENT_WATCHDOG_INTERVAL_MS=250 (new env knob in
server.ts, NaN-safe, production default unchanged) + polls for the
boot line and the tick's stay-alive log — strictly stronger (the old
form never proved a tick observed the parent death). 24s → 3.6s.
- stop-dead-daemon / terminal-agent-owner-watchdog: the 300s/30s
stand-in child lifetimes become stdin-EOF-bound — the child can never
self-exit mid-test on a slow runner (spurious-failure class) and
self-reaps instantly if the test dies (no 300s orphans). Node-compat
stdin APIs (owner-watchdog runs on the Windows lane).
- browser-skill-commands: the sleeper fixture's 30s self-time becomes
8s (no stdin pipe exists in runToFiles) — far above the 1s product
timeout it must outlive, below the test ceiling, so a timeout-kill
regression fails on clean assertions instead of an opaque bun
timeout; added: stdout must NOT contain 'done'.
45/45 green across the four files + server tripwires.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): gen-skill-docs + catalog-trim leave the serial mutator shard
gen-skill-docs.test.ts's 15 in-place generator spawns now render into
mkdtemp out-dirs (gitignored-artifact reads repointed; the handshake
scan's silent console.warn degrade became a hard assertion); its
tracked-tree reads (freshness dry-run, SKILL.md content pins) stay
reads. catalog-trim needed no change beyond the earlier main() guard —
its import is now side-effect-free (pinned by the import-purity test).
Both TREE_MUTATING entries deleted in this commit, per the transition
rule: an entry leaves in the same commit as the file's last in-place
write.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): skill-validation renders codex host into an out-dir
Its 3 in-place --host codex regeneration sites collapse into one
module-level --out-dir render; assertions untouched. TREE_MUTATING
entry deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): host-config self-provisions goldens (ordering dependency severed)
Its goldens were 'produced by gen-skill-docs.test.ts' with a
when-missing beforeAll fallback that wrote the live tree — an
inter-test ordering dependency the serial shard hid. It now renders
codex+factory UNCONDITIONALLY into its own out-dir and reads goldens
only from there (the Claude golden deliberately keeps reading tracked
ship/SKILL.md — a read; out-dir claude renders repoint section-base
paths by design). TREE_MUTATING entry deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): gbrain-detection-override drops mutate-then-git-restore
regenAndSnapshot renders --host claude --out-dir <mkdtemp> (+
--respect-detection) and snapshots probes from the out-dir. The
git-restore machinery is deleted outright — it restored only
PROBE_FILES of the 71 files each call wrote, so a stale tree kept the
other 68 dirty (the partial-restore bug), and its 'no output-path arg'
comment had been false since --out-dir landed. TREE_MUTATING entry
deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): catalog-mode-full renders to out-dir; restore machinery deleted
The full-catalog smoke no longer rewrites all 71 SKILL.md then
regenerates to restore (with its 'CRITICAL: failed to restore' prayer
path) — it renders into a mkdtemp and additionally asserts tracked
ship/SKILL.md is byte-unchanged. TREE_MUTATING entry deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): idempotency proof strengthens to two-out-dir recursive diff
Two renders into two separate out-dirs, EVERY file diffed byte-for-byte
(claude-only and --host all; normalization only for each dir's own
sanctioned section-base repoint; presence-sanity lists guard against a
vacuous empty-dir pass) — strictly stronger than the old in-place
double-regen that sampled 5 files. TREE_MUTATING entry deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): spec-template-sync compares an out-dir render, not an in-place one
TREE_MUTATING entry deleted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): the serial tree-mutating shard dissolves — TREE_MUTATING is empty
Zero mutators remain (all eight render into out-dirs now), so the four
ratchet READERS (parity caps, size budgets, carve parity/ordering) get
a quiet tree by construction in any shard and rejoin the parallel
phase. The ~35-40s serial tail on every full-suite run is gone. The
mechanism stays: a future test that genuinely must write shared
artifacts in place earns an entry with a reason and is serialized
again; the census pin still fails on renamed keys.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(gen): out-dir byte-identity + tree-clean pins for external hosts
codex render: porcelain unchanged AND out-dir gstack-ship/SKILL.md
byte-identical to a fresh in-place render (+openai.yaml presence);
--host all render: exit 0, porcelain unchanged, claude + .agents +
.factory + llms.txt + openclaw docs all present in the out-dir.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): commit the initial free-test durations seed (496 files)
Recorded via --record-durations on a quiescent tree: 479s serial
total, p50 92ms / p90 1.8s / max 31.4s — the top-heavy cost shape LPT
packing exists for. A hint, not a contract: refresh opportunistically
with bun run test:free --record-durations.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(evals): planner/executor/report modes — the CI re-platform surface
One PLANNER computes diff selection + the slice plan ONCE and writes a
manifest (--emit-plan <path> --slices K); K executors consume it
(--plan <path> --slice i), never self-selecting, and write slice-result
artifacts; a REPORT reconciles results against the manifest (--report
<dir>) fail-closed: a slice whose artifact never landed is a FAILURE,
a planned shard nobody reported fails, wrong-slice/duplicate/cross-tier
results fail. Kills per-slice selector divergence and hollow-lane
aggregation at the root.
- hollow-shard guard: under EVALS_ALL, exit 0 with ZERO executed tests
(bun's 'Ran N tests' now captured by the classifier — additive) is
'passed-empty' and fails the run; selective runs keep it 'passed'
with one warning (in-file diff/tier self-skips are legitimate there);
unknown counts are never guessed hollow
- retry parity: --retry 1 default + RETRY_OVERRIDES literals for the
three files whose old matrix rows earned retries: 2 (stale entries
pinned against disk)
- live smoke: gate plan = 48 shards across 6 slices; report mode exits
1 on a fabricated missing slice, 0 when complete
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(ci): sliced paid lane (planner -> 6 executors -> fail-closed report)
The parity-phase re-platform: evals.yml gains a second, sliced lane
driven by scripts/test-paid-shards.ts — the SAME engine local
eval:bg:gate uses, so CI and local share one selection engine.
- plan-slices: ONE planner (fetch-depth 0 — the only job needing
history) emits the manifest; selection fails open to run-all, never
per-slice (the divergence class is structurally dead)
- eval-slices: 6-way matrix consuming the manifest; PTY seed +
skill-registration steps run unconditionally (idempotent — a sliced
lane cannot key them on suite names); aggregate spawn budget
6 x EVALS_JOBS=2 x EVALS_CONCURRENCY=2 = 24 lane-wide (the matrix's
40-way per row queued session startup behind 39 siblings — the
timeout-flake family root); slice results + spooled shard logs
uploaded as artifacts
- slices-report: reconciles slice artifacts against the manifest
FAIL-CLOSED via --report — a slice whose artifact never landed, or a
planned shard nobody reported, is a failure, not an absence
- sequenced needs: evals so provider concurrency never doubles while
both lanes coexist; the matrix + its ratchets are deleted after
demonstrated parity (intersection + expected-additions comparison)
- workflow_dispatch gains evals_all (default true) for parity runs and
post-merge smokes — a dispatch can never silently select zero
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(ci): weekly periodic lane runs EVERY periodic test + gate census backstop
evals-periodic.yml re-platforms onto the sharded runner: planner
manifest → 6 executor slices → FAIL-CLOSED report. This IS the coverage
contract: all ~70 periodic-tier files weekly (EVALS_ALL=1), killing the
silent-rot class where a hard-coded 9-file matrix left ~57 files
running NOWHERE (the autoplan E2E rotted invisibly for months).
- test/helpers/periodic-exclude-data.ts: reasoned exclusions in their
OWN literals file (deliberately not touchfiles-data — map-diff
evaluates old versions of that file standalone). Every entry carries
reason + tracking with a re-entry condition; the runner surfaces each
exclusion per run; policy test pins real-file + non-empty fields.
Initial: ship-idempotency + brain-privacy-gate (documented-red,
never green) and skill-e2e-ios (manual hardware). The TODOS 'sidebar
E2E trio' turned out already deleted — only tombstone tests remain.
- gate-census job: weekly EVALS_ALL gate-tier run — PR lanes are
diff-billed, so without this the full gate census might never execute
anywhere; with the hollow-shard guard it is a census-health check
(exit 0 + zero executed tests fails), not just a test run.
- failure notification is a concrete gh issue UPSERT (one tracking
issue, commented per red week — never issue-per-week spam), with
issues:write scoped to the report job.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: TESTING_INTERNALS covers the 2026-08 runner overhaul
LPT-packed free suite + --record-durations, the emptied TREE_MUTATING
mechanism, the sharded paid runner as the single selection engine,
CI planner/executor/report with the fail-closed report and hollow-shard
guard, the weekly coverage contract + exclusions policy, and the
eval-budgets timeout tiers.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(CLAUDE.md): testing prose matches the overhauled runners
- bun run test: duration-packed shards + --record-durations; the
trailing serial tree-mutating shard no longer exists
- two-tier system: the sliced CI lanes (one engine local+CI), the
weekly all-periodic coverage contract + exclusions, the gate census
- periodic detach timeout 32400 → 37800
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(TODOS): close the absorbed test-infra items, file the overhaul follow-ups
Closed with receipts: the periodic coverage contract (implemented as
full weekly coverage + exclusions), the eval-harness observability P1
(verified already landed: heartbeat, incremental _partial persistence,
live stderr + eval-watch), and the sidebar trio (already deleted —
tombstones remain). Filed: matrix deletion after parity, the
required-check maintainer decision, browse /tmp-namespace hardening,
PTY boot-readiness waits, the single typed test registry, bun-native
LPT swap, runBin/free-runner migrations, eval-list partial exclusion,
phantom key cleanup, duration-weighted slicing.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* v1.73.0.0: test/CI overhaul — green means green, suites restructured for speed
Version + release notes for the audit-and-overhaul branch: every
silently-skipping or never-running test class fixed and tripwired, the
free suite duration-packed with the serial mutator shard dissolved, the
paid lane re-platformed onto the sharded runner (planner/slices/
fail-closed report, parity phase), the weekly all-periodic coverage
contract, eval-budget timeout tiers, and 95 new coverage tests.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): first-live-run fixes — executor history + two environment-blind assertions
The sliced lane's first run (PR #2721) did its job: the planner and
report worked, the manifest governed, and every failure had a name.
Three were fixable on the spot:
- executor + gate-census checkouts get fetch-depth: 0 — files with
SELF-derived selection (the LLM-judge map, routing) walk git at
module load, and selection is deliberately fail-closed on git errors,
so the shallow checkout crashed those shards ('ambiguous argument
main...HEAD'). The manifest still governs WHICH shards run.
- landscape --toc gate: the exact toBe(3) landscape-page count was
font-metric-dependent (3 on Amazon Linux, 2 on ubuntu CI — the same
disease the file's own page-index comment warns about). Now a
comparative invariant: --toc must not CHANGE the landscape count vs
a baseline render.
- paid-run-manifest parse test builds its manifest under EVALS_ALL so
it never walks git (proven with GIT_DIR=/nonexistent).
Remaining first-run failures are newly-exposed rot in gate files that
had never executed in CI (skillify D1 refusal, session-intelligence
context-restore, one tpa-apple-ban retry flake) — being probed
separately; they are the lane WORKING, not the lane failing.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(TODOS): file the three first-execution findings from the sliced lane's live run
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* v1.74.0.0: queue-advance — #2722 claims the v1.73.0.0 slot
The version gate caught a live queue collision (its whole job); same
MINOR bump level, next free slot per bin/gstack-next-version.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): per-shard CHROMIUM_PROFILE — the collision class duration packing exposed
Nine test files launch in-process persistent contexts or daemons that
default to the SHARED ~/.gstack/chromium-profile. Two concurrent shard
processes on one profile dir kill each other's browser — observed live
on CI once duration packing recomposed shards: handoff's
launchPersistentContext died 'Target page, context or browser has been
closed' (--user-data-dir=~/.gstack/chromium-profile in the call log)
while a sibling shard's daemon logged 'Chromium process crashed'. Hash
sharding had masked the collision by chance placement; handoff passes
standalone everywhere.
Fix at the runner, not per file: each shard child gets
CHROMIUM_PROFILE=<shard-state>/chromium-profile (the documented env
knob, same isolation idea as the existing per-shard TMPDIR). Files
within a shard run serially, so sharing the per-shard profile is safe;
config.test's resolution-order tests save/restore the env around their
assertions.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): landscape --toc gate asserts promotion PRESENCE, not counts
Two rounds of CI receipts: the exact toBe(3) was font-metric-coupled
(3 on Amazon Linux, 2 on ubuntu), and the baseline-comparison repair
then failed 2-vs-3 across renders SECONDS apart in one CI job while the
sibling no-toc test saw 3 — per-render image-promotion timing makes any
count assertion here a coin flip. The sibling test owns exact promotion
counts; this test's actual invariant is that --toc does not break the
promotion machinery: >=1 landscape page + the TOC rendered. Also drops
the second render (halves the test's runtime).
Flaky per-render image promotion itself is worth its own look — noted
in TODOS with these receipts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(TODOS): file the per-render image-promotion nondeterminism (receipts from PR #2721)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): per-FILE Chromium profiles for the nine in-process launcher files
Completes the profile-isolation work: the per-shard CHROMIUM_PROFILE
stopped cross-shard kills; these nine files launch in-process
persistent contexts and could still collide with a lingering daemon a
sibling file spawned on the SAME shard profile. Each now scopes a
mkdtemp profile via beforeAll/afterAll (the module-scope-tripwire-safe
pattern), cleaned up per file. All nine green solo and in combined
runs, except the pre-existing commands+snapshot pairing — proven
identical WITH and WITHOUT these edits (baseline receipts) — which is
the daemon-lifecycle follow-up now extended in TODOS with this
session's receipts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(browse): Chromium-crash exit is daemon-only — embedded launches never kill their host
handleChromiumDisconnect unconditionally process.exit()ed. Correct for
the standalone daemon (its supervisor/user must notice); suicidal when
a TEST launches BrowserManager in-process: a mid-suite Chromium death
exited the whole bun shard with no terminal summary — the exact
truncation class the strict runner flags (observed live: CI shard 1 on
eb233299 died at cache-concurrent-refresh right after a daemon-spawning
gate test; with this fix the same pairing runs to completion and
REPORTS instead of dying).
The standalone entrypoint opts in via markDaemonProcess() under
server.ts's import.meta.main gate — the same embedder contract its
signal handlers already use (gbrowser phoenix keeps its own handlers).
Embedded contexts now get the disconnect log line and continue.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): context-restore assertion is evidence-based, not prose-matching
The test failed twice per run in TWO CI cycles while passing locally
4/4: the prompt said 'present the content' and the check grepped the
FINAL message for exact phrases — local runs quoted the file, CI runs
paraphrased ('the most recent context is from branch-b...') and the
substring check lost the coin flip.
- prompt now demands machine-checkable output: the newest file's
'## Working on:' heading VERBATIM + a literal 'RESTORED: <filename>'
marker (the mtime-scramble and cross-branch subject matter untouched)
- assertion ordered strongest-first: RESTORED marker → legacy content
phrases → tool-call corroboration (Read/Bash input naming the newer
file, credited ONLY when the older file was never read — a
both-files run must still present the right one)
- the older-file negative got STRONGER: an explicit RESTORED marker
naming the older file fails even if wintermute words appear elsewhere
- sibling scan: context-recovery-artifacts got the additive prompt-side
treatment only (quote the matched literals verbatim); its lenient
1-of-6 assertion deliberately unchanged
3/3 consecutive local green with all evidence classes firing
(marker=true, content=true, toolNewer=true, toolOlder=false).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): skillify family — HOME==cwd broke project-skill registration
Root cause (forensically pinned from stream-json init events + a
kill-after-init probe): with HOME set EQUAL to the child's cwd, claude
resolves <cwd>/.claude/skills as the PERSONAL skills directory and the
seeded project-tier skills never register — the Skill tool returned
'Unknown skill'. The provenance-refusal test then improvised a refusal
whose wording missed the regex (the deterministic CI+local red); the
happy-path and approval-reject siblings passed only because their
agents self-recovered by Reading SKILL.md manually — silently not
exercising the Skill-tool path at all.
All three tests now use HOME=<workDir>/home (a fresh subdir keeps the
override's intent: child ~/.gstack writes land in the assertable
sandbox, without the cwd collision). Refusal test additionally: a
'not registered/unknown skill' tripwire (a not-loaded skill can never
pass as a refusal) and the refusal regex now matches assistant text
only — the skill BODY echoed into the transcript contains the exact
refusal message, so the old full-surface match could pass vacuously
once the skill loaded. Sibling disk assertions sweep both $HOME/.gstack
and cwd .gstack roots (positives and negatives).
Verified paid: refusal 2x consecutive green with the skill's EXACT
message rendered ('Launching skill: skillify' in-transcript), then the
full file 5/5 green (~$1.35) with both siblings driving real Skill
calls (25-27 turns each).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(TODOS): two of three first-execution findings fixed (skillify family, context-restore)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): context-restore gets a private home — the REAL root cause was fixture sharing
The evidence-based assertion fix was treating a symptom. The slice
artifact's embedded transcript showed the CI agent restoring
20260829-context-save-skill-test.md — the checkpoint the SIBLING
context-save test wrote into the SHARED gstackHome checkpoints dir,
which by filename-prefix ordering genuinely IS the newest. The agent
behaved CORRECTLY; the test's fixture set was open to concurrent
sibling writes, and bun --concurrent ordering differs between CI (save
finished first) and local (restore listed first) — the entire
local-green/CI-red split explained.
The restore test now uses its own .gstack-restore-home (the whole home
moves, not just the handed path — an agent deriving the dir from
GSTACK_HOME/projects/<slug> must land in the closed set too). Full file
4/4 paid green with all evidence flags firing.
Also: the on-failure shard-log artifact glob uploaded nothing — the
Fix-bun-temp step points TMPDIR at /home/runner/.cache, so the spool
lands there, not /tmp. Both eval workflows now glob both locations
(this gap is why diagnosing THIS failure required digging transcripts
out of the slice-results artifact).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(evidence): carry the real index mtime onto gstack-wtree's temp copy
The stat-cache seed (cp of the real index) stamped the temp index "now",
which defeats git's racy-git protection: an entry is only re-hashed when
its cached mtime is not older than the index file itself, so a same-size
rewrite landing in the same second as the last real index write looked
non-racy, kept its stale stat-cache entry, and vanished from the
fingerprint — evidence stayed FRESH after a source change. This is the
CI flake in test/evidence.test.ts "allow-paths carve-out" (sub-second
alignment on fast runners: expected STALE exit 1, got FRESH exit 0).
touch -r restores the original index timestamp, reinstating the exact
racy window git itself uses. Deterministic regression pin in
test/review-log.test.ts reproduces the miss with pinned zero-nsec
timestamps (fails on the old script, passes now); receipts: manual
probe shows the fresh-stamped copy returning the clean tree for a
same-size 'hello'→'howdy' rewrite while the mtime-carried copy detects
it.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): landscape gate bounds the promotion count instead of pinning 3
The alt-hinted image promotion rides the per-render measurement race
already filed in TODOS (2-vs-3 landscape pages on renders seconds
apart — CI receipts from PR #2721, now reproduced locally). Pin the
two deterministic promotions as the floor and the three promotable
blocks as the ceiling (anything above 3 means the veto leaked); the
veto/portrait assertions remain exact.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: Test <test@test.com>
* feat(gen): strip gen-time-only frontmatter keys from Claude renders
interactive + benefits-from are read from the .tmpl by buildContext at
generation time; no runtime, host, or test reader consumes them from the
generated SKILL.md (e2e-harness-audit reads .tmpl; benefits-from tests
assert rendered prose). gbrain: stays (bin/gstack-brain-context-load reads
it from the installed render); hooks: stays (Claude Code host wires
PreToolUse from it).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(gen): regenerate SKILL.md — dead frontmatter keys removed
Mechanical regen after hosts/claude.ts stripFields change.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(test): context-budget ratchet — CI ceilings on always-on + eager token ledgers
New free test grades the two ledgers nothing else guards: the full-frontmatter
always-on catalog (aggregate) and per-skill eager tokens (SKILL.md +
forced-read refs), via checkBudget from lib/context-bill.ts. Ceilings live in
test/fixtures/context-budget.json with x1.05/x1.10 headroom; regenerate with
bun test/helpers/capture-context-budget.ts. New skills fail until consciously
budgeted; removed skills fail until the fixture is refreshed; reductions
ratchet the ceilings down so wins lock in.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(todos): file output-template carve wave + plan-ceo doctrine revisit; mark preamble-carve P3 in flight
Two follow-ups deferred from the approved token-reduction program (CEO review
'NOT in scope' list), filed with full context per TODOS format. The existing
P3 preamble-carve entry gets a status update pointing at the program that
supersedes it.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): review findings — Windows path normalization, full totals rebuild, ratchet coverage
Pre-landing review (5 specialists) found one critical: the ratchet test runs
in the curated Windows lane, where path.relative yields backslash skill names
that miss the test/ filter and mismatch every POSIX fixture key. Names are now
normalized once in buildRatchetBill (toPosixName) and the fixture filter is
tightened to test/fixtures/. All eight Bill.totals fields are rebuilt from the
filtered list (no fixture-polluted perInvocation/totalMd numbers for future
consumers). New coverage: Windows-separator normalization pins, a
captureContextBudget round-trip against tree-a (headroom math exact), a
stripFields regression pin (interactive/benefits-from absent from renders,
hooks/gbrain preserved), and the ceilings test no longer double-reports
stale-fixture entries.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): adversarial findings — stable root key, symlink-alias dedupe, fixture-shape guard
Adversarial review (Claude subagent) verified the fixture's root-skill key was
the capture machine's checkout dirname: any non-gstack-named clone (every
Conductor worktree) failed the free suite, and the documented re-run-the-capture
recovery baked the local dirname into the committed fixture — silent corruption
through the tool's own protocol. The root skill is now pinned to ROOT_SKILL_KEY
('gstack', its frontmatter name). Symlink aliases are realpath-deduped (census
precedent): connect-chrome no longer gets its own ceiling, so Windows checkouts
that materialize the symlink as a plain file can't fail the stale-ceiling
set-equality test. New guards: fixture-shape validation (a string alwaysOnTotal
can no longer silently disable the ceiling), a mutation pin that the filter
shrinks the always-on ledger vs the raw bill, an alwaysOnTotal violation test
(the branch was load-bearing with only under-budget coverage), and an atomic
temp+rename fixture write. Fixture regenerated: 59 ceilings, alwaysOnTotal 6344.
Deferred with a TODO: anchoring transformFrontmatter's denylist strip to the
frontmatter block (latent, zero live collisions, pre-existing path).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: bump version and changelog (v1.69.1.0)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: update project documentation for v1.69.1.0
CLAUDE.md: Token ceiling section documents the context-budget ratchet as
the third guard (test file, fixture, new-skill budgeting, capture command).
CONTRIBUTING.md: Tier 1 guard list gains a Context-budget ratchet bullet;
the Adding-a-new-skill checklist gains the budget-capture step.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: pin exact guard semantics for the context-budget ratchet in CLAUDE.md
Doc-review finding: "a third enforced ceiling" undercounted the guard
family (skill-size-budget floors and parity ratios also watch these
ledgers, relatively). Rephrased to match the ratchet test's own header:
absolute ceilings vs relative floors/ratios.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(changelog): heaviest-skill claim matches the fixture (land-and-deploy edges review by 0.2%)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(bin): gstack-skill-start + gstack-skill-end — the preamble runtime, consolidated
Absorbs the ~13KB of bash every tier-2+ SKILL.md inlined twice over (bootstrap
fence + artifacts-sync fence) and the skill-end telemetry/sync fences. Same
KEY: value STATUS-line contract the prose interprets, plus SKILL_START_PROTO
handshake (OV5), SESSION_ID/TEL_START echoes, GSTACK_HOME-normalized state
paths (EOV7), --parent-pid session identity (EOV5: $PPID inside the script is
the ephemeral tool-call shell), OV4 sanitization of passthrough output, and a
receipted daily artifacts pull (_receipted_git, brain-sync class, fail-closed).
Per-line || true error style throughout (F3) — a mid-script failure never drops
later STATUS lines.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(gen): preamble resolvers emit a script invocation fence instead of inline bash
generate-preamble-bash: ~6.3KB fence -> 4-line gstack-skill-start invocation
(quoted-tilde pitfall handled: leading ~ interpolates through $HOME; env-var
hosts keep $GSTACK_BIN) + degraded-mode prose (F1/EOV8: safe defaults, consent
gates deferred-never-lost; OV5: proto rule). generate-brain-sync-block: ~6.8KB
bash -> interpretation prose + the privacy stop-gate (stays inline until
Phase 2's gated emission). generate-completion-status: telemetry fence -> one
gstack-skill-end call with SESSION_ID/TEL_START handoff.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(gen): regenerate all skills + golden fixtures — inline preamble bash removed
Mechanical regen after the resolver change: −12,628 lines across 52 renders
(corpus 952K -> 806K render tokens; tier-2 skills −11-13KB each). Golden
per-host ship fixtures refreshed from the fresh claude/codex/factory renders.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: skill-start contract suite + preamble A/B eval + touchfiles registration
test/gstack-skill-start.test.ts (11 free tests): STATUS-key contract vs the
prose (F2), per-host fence resolution shapes (E1), proto-first, OV4 marker
sanitization, --parent-pid identity, headless suppression, skill-end duration
math + pending cleanup. test/skill-e2e-preamble-script-ab.test.ts (gate tier,
OV7): inline-bash render (pinned from 29785978) vs script render with the
fence redirected at the worktree bin (EOV2 — hermetic evals otherwise resolve
the operator install and silently exercise degraded mode). 21 touchfiles dep
lists gain the two bin scripts (EOV9) so future script edits select the
preamble evals; selection-count pin updated 23->24.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: repin ~70 assertions to the script contract — every literal gets a successor
Assertions that pinned inline-bash internals (update-check guard, _SESSIONS
reaping, telemetry start/end blocks, routing probe, repo-strip producer,
first-task gating, EXPLAIN_LEVEL/QUESTION_TUNING echoes, #2499 jq scope
resolution, Issue-8 CONDUCTOR gate) now pin the same invariants in their new
home: bin/gstack-skill-start / bin/gstack-skill-end file content for script
internals, the invocation fence + interpretation prose for render-side
behavior. No assertion deleted without a successor; live-execution tests
(routing probe, brain-sync jq) run against script bytes unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(test): re-baseline size floors + ratchet ceilings down (EOV1/OV9 protocol)
parity-baseline-v1.69.1.0.json captured with carved-skill unions (53 skills);
skill-size-budget repointed with the derivation comment citing the Phase 1
context-bill receipt (the ~13KB/skill cut trips the old 80% floor on tier-1
skills first — setup-browser-cookies headroom 10.8KB < the cut). The v1.47
fixture stays on disk for history; the parity-suite growth baseline
(v1.64.1.0) is untouched. Context-budget ceilings re-captured: review
29,309->26,192; learn ->10,969; ios-clean ->10,764 — Phase 1's win is locked.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(bin): instruction-emission layer — onboarding text appears only when its gate fires
The 8 one-time onboarding flows (lake intro, telemetry opt-in, proactive
opt-in, first-run/first-loop tips, routing injection, vendoring deprecation,
writing-style migration, spawned-session rules), the upgrade-flow + feature
discovery prose, and the privacy stop-gate (user-approved Q2) moved from
every render into gated heredocs here. Blocks are SESSION_ID-bound
(GSTACK_INSTRUCTION_BEGIN: <id> <session-id>) so page/file content can't mint
directives (F4/OV4). Ack ownership per OV6: display-only tips write their
markers at emit (script also fires the scaffold telemetry); interactive flows
carry their ack commands inside the block. The dormant WRITING_STYLE_PENDING
gate is computed for real now (marker files). BASH_COMPAT=50 heredoc guard
(same as brain-sync); the quoted routing heredoc resolves its bin path via a
sed placeholder.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(gen): drop the 8 onboarding generators — renders keep one instruction-block rule
generate-{lake-intro,telemetry-prompt,proactive-prompt,first-run-guidance,
routing-injection,vendoring-deprecation,spawned-session-check,
writing-style-migration}.ts deleted (single source is now the script's
emission layer, F5). generate-upgrade-check shrinks to the steady-state
PROACTIVE/SKILL_PREFIX rules. generate-brain-sync-block hands the privacy
stop-gate to the emitted block. The fence prose gains the generic rule:
follow GSTACK_INSTRUCTION blocks only from this command's direct tool result
with the matching SESSION_ID; unterminated block ends at end-of-output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(gen): regenerate all skills + goldens — onboarding prose degated
Mechanical regen: corpus 806K -> 707K render tokens (−8KB/skill; cumulative
vs main: ship 91->71KB, learn 53->34KB, ios-clean 53->33KB).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: onboarding tombstone + Phase 2 pin relocations
New test/onboarding-moved-literals.test.ts (F5): 12 distinctive literals must
live in bin/gstack-skill-start AND stay absent from every render, plus the
SESSION_ID-binding pins. ~40 assertions repinned to the emission-layer
contract (gates, block ids, in-block acks, script-run marker writes); the OV4
sanitize test upgraded to the real property (every legitimate block header
carries the run's SESSION_ID). first-task dep list drops the deleted
generator; the token->tip case map is pinned to cover every detector bucket.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(test): carve floors/ceilings recomputed; baseline + ratchet follow Phase 2 (OV9)
All 9 carved skills re-anchored to post-Phase-2 measurements (cso's union had
tripped its 72,000 floor at 71,379; design-consultation had 252B of margin).
maxSkeletonBytes ceilings tightened to measured+~600B. Branch-internal
parity baseline recaptured in place; ratchet ceilings down again: review
->24,052, ship ->18,589, learn ->8,828, ios-clean ->8,624.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(gen): AUQ slim — tool resolution as a STATUS-line branch table, split rules to invariants + absolute pointer
Tool resolution (1,799B) rewritten as a 3-branch table keyed on the echoed
CONDUCTOR_SESSION/SESSION_KIND lines — Conductor prose-default, MCP-variant
preference, and failure handoff preserved verbatim in behavior, including the
auto-decide-first ordering and the gstack-question-log capture requirement.
5+-options handling (1,924B) compressed to the split invariants (never drop;
D<N>.k shape; Include/Defer/Cut/Hold; question_id scheme with the never-ask
refusal) + the full-rule pointer. Both doc pointers now interpolate the
absolute install root (Codex outside-voice #7 convention) instead of the bare
'in the gstack repo'. Failure-fallback, Format, and self-check sections are
byte-identical — all 14 MANDATORY always-loaded pins pass with zero test
edits.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(gen): regenerate all skills + goldens — AUQ slim
Mechanical regen: −1.3KB per tier-2+ skill (ship 69.9KB, learn 32.5KB).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(test): baseline + ratchet follow Phase 3 (OV9); OV8 evaluated — shrink floor stays
Branch-internal baseline recaptured; ratchet ceilings down again. OV8's
floor-retirement question, evaluated as planned after Phase 3: the 80% shrink
floor stays — it uniquely catches accidental body deletion in non-carved
skills BETWEEN ratchet recaptures, and the capture command has amortized the
fixture-refresh cost that motivated retiring it.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(review): carve adversarial, plan-completion, and review-army into sections
The three resolver macros ship already carves as siblings now load on demand
for /review too: skeleton 100.2KB -> 55.0KB (-45%), union 93.4KB. Resolvers
stay the single source of truth (sections wrap the macros). Step 0/1, scope
drift, critical pass, confidence calibration, and fix-first stay always-loaded.
Fixtures and pins follow the moved content (codex-hardening wrapped-sites,
review-army E2E fixture builds skeleton+sections with an empty-fixture guard).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(codex): carve the three mutually exclusive modes into sections
Review/Challenge/Consult mode bodies (34.7KB where at most one ever runs)
load on demand: skeleton 81.0KB -> 55.2KB, union 1.04x the monolith. The mode
dispatch, filesystem boundary, and a new always-loaded 'Synthesis
recommendation (REQUIRED) — all modes' block stay skeleton-side (the AUQ
per-skill pins pass unchanged); the plan-file report + exit gate render after
the last section pointer per the gateAfterStop pattern.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(land-and-deploy): carve first-run validation, readiness gate, and merge/deploy into sections
The once-per-repo dry-run validation, the pre-merge readiness gate, and the
merge + deploy-strategy steps (37.8KB) load on demand: skeleton 91.1KB ->
55.7KB. Step 1.5 keeps its detection bash as the dispatch; the first-run
section's fingerprint-save block gained {{SLUG_EVAL}} so it is self-contained.
Zero content lost (line-coverage checked against HEAD).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(ios): demote the four ios skills to preamble-tier 2 (Phase 5)
They never consume the tier-3 sections (repo-mode ownership, search-before-
building) but do fire AskUserQuestion, which tier >=2 provides — verified by
grep before the plan review. -2.2KB per skill. Render assertions pin the
demotion (tier-3 sections absent, AUQ format present).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(guards): register wave-1 carves; monolith invariants retire; baselines + ratchet follow
CARVE_GUARDS gains review/codex/land-and-deploy (12 carved skills total);
their MONOLITH_INVARIANTS entries retire (invariants now generate from the
registry, cso precedent). Touchfiles: carve-section-loading covers the three
new carves; the codex + land-and-deploy LLM-judge dep lists widen to their
sections. Regen + goldens + branch-internal baseline + ratchet ceilings
recaptured (review 24,052 -> skeleton-based ceiling; union floors hold).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(gen-skill-docs): review render pins read the carved union
The review carve's readSkillUnion conversions (same pattern its neighbor
carved-skill pins already use).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(autoplan): carve the four review phases + tasks aggregator into sections
Phase bodies (CEO/Design/Eng/DX consensus flows) and the Implementation Tasks
aggregator load on demand; Design and DX stay separate sections because each
is independently conditional on scope. Skeleton 83.7KB -> 58.7KB (-30%
always-loaded); the 6 decision principles, classification, sequencing, and
explicit skip-condition dispatch stay always-loaded. The chain E2E's
phase-complete markers now live only in sections, so its assertions double as
section-read proof (behavioral: external).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(spec): carve the post-confirmation gate-and-file tail into one section
Phases 1-4 are the turn-1 conversational spine — carving them would force the
Read on the first user message for zero real savings. The mechanical tail
(4.5/4.5a/4.5b redaction gates + Phase 5 filing + TTHW telemetry) fires only
after draft confirmation: a genuine lazy boundary, kept as ONE section so the
gh-issue-create bash can never load without the fail-closed redaction gate
that precedes it. Skeleton 65.4KB -> 50.7KB; all ~85 phase-structure
invariants migrated location-aware plus a new carve-shape suite (56 tests).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(setup-gbrain): carve the branch-exclusive install paths into sections
Brain-init (Paths 1/2/3/4 bodies), engine remediation, transcript gate, and
CLAUDE.md persist load on demand — at most one install route ever runs.
Skeleton 75.3KB -> 57.0KB; the Step 1 detect and Step 2 path dispatch stay
always-loaded. New buildSetupGbrainFixture helper gives the periodic E2Es
extract-don't-copy fixtures with a non-empty guard; the voyage-code-3 gate
counts scan the tmpl union (the third init site lives in engine-remediation).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore(guards): register wave-2 carves (15 carved skills); autoplan monolith retires; baselines follow
CARVE_GUARDS gains autoplan (behavioral: external via the chain eval), spec,
and setup-gbrain; autoplan's MONOLITH_INVARIANTS entry retires. Touchfiles:
setup-gbrain periodic dep lists gain the section tmpls + fixture helper; the
stale-brain-refs scan covers setup-gbrain/sections. Regen + goldens + branch
baseline + ratchet recaptured.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(qa): carve QA patterns + health rubric into on-demand sections (68→48KB skeleton)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(browse): carve full command list + snapshot flags into sections/command-list.md (39→27KB skeleton)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(retro): absorb inline git/awk metrics into bin/gstack-retro-metrics + carve report format
RETRO_METRICS_PROTO: 1 contract, local git reads only (fetch stays in the
skill prose), degraded path documented in the skeleton.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: register wave-3 carves (qa, browse, retro) — guards, touchfiles, pins, baselines
CARVE_GUARDS gains the three entries; qa's monolith invariant retires.
auq-format carve-safety now keys on the skeleton+sections union shipping
the AUQ block (first tier-1 carve: browse never renders it by design).
Baselines: parity v1.69.1.0 at 18 sectioned skills; ratchet recaptured.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): drop stale generate-lake-intro import (generator deleted in the emission-layer move)
Sol scope discipline stays pinned via the model overlay + completeness
section; the lake intro is now a single script-emitted blurb.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(office-hours): carve Phase 2A/2B into mode-exclusive sections (81→67KB skeleton)
A session runs exactly one mode, so a builder session never loads the
13KB startup diagnostic. Mode mapping and the vibe-shift upgrade rule
stay in the skeleton.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat(design): carve UX doctrine + Pretext patterns into read-on-demand sections
design-html 57→49KB, design-shotgun 53→50KB. Sections wrap
{{UX_PRINCIPLES}} so scripts/resolvers/design.ts stays the source of
truth; the pretext-patterns STOP sits at the top of Step 3 so the read
provably precedes the Write.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: register wave-4 carves (office-hours ext, design-html, design-shotgun) — 20 carved skills
Both design entries carry requiredReads + loading-eval scenarios (D3A
condition). office-hours phase sections are mode-exclusive, so only the
always-reached design/handoff section is a deterministic requiredRead.
Baselines and ratchet recaptured.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: trim CLAUDE.md 66.4→44.9KB — verbatim moves to docs/, pointers stay inline
Moved: browser/sidebar/server internals, CHANGELOG release-summary format
spec, project tree, hermetic-E2E detail, slop-scan reference, OpenClaw
publishing. Kept inline: every hard behavioral rule (dist/ ban, redaction
scan-at-sink, egress receipts, bisect commits, eval detach, CHANGELOG
entry rules), the machine-managed GBrain block (byte-identical), and the
'## Deploying to the active skill' header with gbrain-refresh in range
(pinned by test/gbrain-refresh-install-render.test.ts). No voice rewrites.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): seed onboarding markers into the hermetic child GSTACK_HOME
EOV7 made bin/gstack-skill-start honor GSTACK_HOME, so the operator-HOME
seeding in e2e-helpers.ts no longer reaches hermetic children — the
emission layer fired lake-intro/telemetry prompts that burned turns and
stalled PTY tests waiting on an answer (observed: plan-mode-no-op derailed
by the telemetry question). Onboarding-specific tests pin their own
GSTACK_HOME per-test, which merges over this seed.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: raise carve-section-loading wall clock to 480s SDK / 540s bun
The heavy full-workflow scenarios satisfy their required section reads
inside 60s but need 300-450s to finish the report on slower sandboxes;
the 300s default read as a loading failure when the carve invariant held
(traces: plan-eng-review read its section at 8s, office-hours all three
at 24s, design-html both at 50s — all timed out mid-report).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(security): harden the skill-start trust boundary — review-army findings
Session ID gains a urandom suffix (block binding unforgeable by reflected
content); _sanitize also neutralizes spoofed SESSION_ID: lines; branch
names are charset-clamped before JSON embedding (skill-start + skill-end);
.brain-last-push reads first line only with a charset clamp; the artifacts
URL echo routes through _sanitize; the privacy consent gate fires in
interactive sessions only (spawned auto-choose could accept consent no
human gave — emission order is not a safety property); the daily pull gets
non-interactive + slow-network git guards and stamps only when the
receipted path ran; ~/.claude.json gets a grep pre-filter before the jq
parse.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(resolvers): question-log session_id becomes a substitution placeholder + stale-comment sweep
The question-log block bound $_SESSION_ID, a shell variable the
consolidated fence never sets — hook-less hosts logged empty session_id,
breaking /plan-tune per-session grouping. It now uses the same
substitute-from-the-skill-start-echoes contract as the telemetry block.
Also: retired the pre-Phase-2 stop-gate docstring, repointed the
gbrain-local-status cross-reference at the script's inline jq, dropped an
orphaned section comment, documented retro-metrics' suffix-only census.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: regenerate renders for the question-log placeholder; goldens + baselines follow
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: hermetic update-check, onboarding gate sequencing, seeding parity
The contract test's child did a live git ls-remote + curl to github.com on
every bun run test (update_check config now gates it off); the headless
test gets a fresh GSTACK_HOME so the suppression is actually exercised; a
new OV6 test drives the script three times to pin ack-at-emit and gate
sequencing; hermetic seeding covers the config-keyed privacy gate; the
EVALS_HERMETIC=0 debug seeding reaches marker parity.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(ci): demote the preamble A/B to periodic (OV7) and add it to the periodic matrix
Post-Phase-3 demotion per the plan; the eval needs fetch-depth 0 (it git
shows a pre-Phase-1 sha), which only the periodic workflow provides — and
a static matrix entry so it can't silently never run.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: bump version and changelog (v1.70.0.0)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: update project documentation for v1.70.0.0
ARCHITECTURE.md: the preamble section now describes the v1.70 runtime —
the rendered {{PREAMBLE}} block invokes bin/gstack-skill-start and reads
STATUS lines, gstack-skill-end logs telemetry, and one-time onboarding
text arrives as gated GSTACK_INSTRUCTION blocks instead of riding in
every render.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: doc-review fixes — repair moved-file links, drop unbacked session-count claim
docs/BROWSER_INTERNALS.md: the two ARCHITECTURE.md anchor links broke when
the section moved from repo-root CLAUDE.md into docs/ — now ../ARCHITECTURE.md.
ARCHITECTURE.md: the preamble's session-tracking item claimed an active-session
count and an "ELI16 mode" that no shipped code implements (the count
computation was deleted with the inline preamble); describe the real
touch-and-prune behavior instead.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs(changelog): correct numeric claims against measured counts
50 of 62 installed skills dropped (fixture/alias entries have no preamble);
11 new carves + a deeper office-hours carve = 9→20; test counts match the
files (13 / 11 / 3 / 7).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: repoint the preamble-runtime version reference after the queue rebump (v1.71.0.0)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(e2e-design): widen the Aesthetic synonym set — vocabulary variance, not a regression
Both attempts in run 33090283032 produced judge-praised DESIGN.md files
phrased as 'design principles'/'design language' without any of the four
original literals; inputs were identical to the prior passing run
32899975845 (design-consultation untouched by the intervening merge).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test): stage design-consultation's sections/ into the E2E fixture
The skill has been carved since v1.57.0.0 — the DESIGN.md structure
prescription (the AESTHETIC proposal template) lives in
sections/proposal-and-preview.md behind a STOP-read. The fixture only
copied SKILL.md, so the agent improvised structure from the skeleton and
the section-synonym check has been a coin flip since the carve (CI run
33090283032 trace shows 'no sections dir'; the local eval store has the
same failure on 2026-08-25 while that day's CI run passed on lucky
vocabulary).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* ci: bump CI image Bun 1.3.10 -> 1.3.13
Matches the local toolchain and brings native `bun test --shard=M/N` /
--parallel to CI (needed by the free-test lane and shard runner work).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: stop version bumps rebuilding the eval Docker image (cache key trio)
Three coupled fixes, atomic because any subset is worse than none:
1. Image tag keys on hashFiles(Dockerfile.ci, bun.lock) — package.json is
out: its version field changed on 60/60 recent commits, forcing a ~2min
image rebuild per PR for a dependency set only bun.lock determines.
2. ci-image.yml now pushes that same content-hash tag (previously only
:latest/:sha, so the weekly prebuild never warmed the tag the eval
matrix actually looks up) and both eval workflows get registry layer
cache (cache-to export gated to same-repo runs; fork tokens cannot
write GHCR).
3. Dockerfile bakes /opt/node_modules_cache/.bun.lock and the runtime
Restore-deps guard diffs bun.lock instead of package.json — otherwise
every version-only bump made all 14 matrix jobs fall back to a live
bun install, which is slower than today's behavior.
Worst-case failure mode is self-healing: a missing tag or cache falls
back to exactly the previous rebuild-and-install path.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: stop double-running lint + skill-docs on every PR commit
Both fired on unrestricted push AND pull_request, so each PR push ran
them twice (12 duplicate (headSha, workflow) pairs in the last 200 runs).
push is now main-only; pull_request covers PR branches.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: run actionlint from the prebuilt image (16s -> ~2s)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: right-size five single-core jobs to ubicloud-standard-2
actionlint, skill-docs, version-gate, pr-title-sync, and the evals report
job never exceed one core; standard-8 was ~4x the cost for zero wall-clock.
build-image and the eval matrix keep standard-8.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: fix workflow_dispatch concurrency collisions (head_ref || run_id)
head_ref is empty on workflow_dispatch, so every manual dispatch of these
four workflows shared one empty-suffix group and cancelled each other.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci(windows): cache bun installs; run the curated suite, not a hand list
- actions/cache on ~/.bun/install/cache keyed on bun.lock (install was
35-45s of both 55-64s jobs, all network) and Bun pinned to 1.3.13 to
match the other lanes.
- windows-free-tests now runs `bun run test:windows` (the runner's
--windows-only curation) instead of a hand-listed 13-file subset that
had drifted from the registry it sampled. POSIX-bound tests get
excluded in ONE place (the curation patterns), not two.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: retry 1, not 2, on every paid path
Measured on the llm-judge shard: --retry 2 amplified 25 tests into 46
executions (+84%), with retried runs at 138s vs a 10-12s baseline (429
backoff), and a permanently-failing test paying 3x. One retry still
absorbs one-off flakes; chronic flakes become visible fix-work instead
of silent wall-clock.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: split skill-e2e-review into three per-file CI shards
Bun runs describe blocks as concurrency barriers, so the e2e-review CI
job executed its tests serially: 741s of an 860s PR critical path for
tests whose slowest member is 224s. The per-file matrix is the repo's
parallelism unit, so the split moves:
- Retro E2E + retro-base-branch -> test/skill-e2e-retro.test.ts
- review/ship base-branch + Review Dashboard Via Attribution
-> test/skill-e2e-review-attribution.test.ts
- sql-injection / enum-completeness / design-lite stay in
test/skill-e2e-review.test.ts
One 741s job becomes three ~180-250s jobs. Locally the worst paid shard
drops from 1705s (94.7% of the 1800s kill) to under 700s. Test names,
bodies, suite strings, and eval-store collectors are unchanged, so
baselines carry over. Matrix rows added to both eval workflows
(attribution is gate-only, so no periodic row); the report job's
hardcoded runner count is gone (drift-proof).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: gate security-bench on SECURITY_BENCH=1, not model-cache existence
The existsSync gate ran ~12s of ONNX inference (plus a HuggingFace
dataset fetch) on every free-suite run on any dev box that had ever
warmed the classifier, while CI (no cache) silently skipped it. Now
explicit opt-in: SECURITY_BENCH=1 bun test browse/test/security-bench.test.ts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: watchdog E2E in 1.5s instead of 22.7s (tunable poll interval)
server.ts gains BROWSE_WATCHDOG_INTERVAL_MS (floor 50ms, default 15s
unchanged). The #994 stay-alive test runs a 250ms tick and waits for the
stay-alive log line instead of blind-sleeping 2s + 20s past the
production interval.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: dedupe coverage gates; route both walks through skill-census
skill-coverage-floor duplicated two matrix assertions (registry
completeness, gate-tier floor) with a DIFFERENT hand-rolled directory
walk — matrix's skipped nothing, floor's skipped node_modules/docs/test.
Two 'same' gates disagreeing on the census is the bug class
test/helpers/skill-census.ts was written to kill. Registry assertions
now live in matrix only (with floor's better error message), both files
walk via skillCensus().authoredSkills, and floor keeps the per-skill
structural checks it owns.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: EVALS_JOBS for shard processes; explicit within-shard concurrency
EVALS_CONCURRENCY was overloaded: the legacy bun-test path used it as
--max-concurrency (default 15) while the sharded runner read it as the
process count — exporting the legacy value gave 15 concurrent Bun
processes each spawning claude (the 429 storm). Now: EVALS_JOBS = shard
processes (default 4); EVALS_CONCURRENCY = bun --max-concurrency inside
a shard (default 4, explicit in shard args — omitting it made
within-shard parallelism silently differ from the legacy path). Stale
49/59 header math replaced with the live-count rule.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: enforce detach-timeout floor from the live shard census
New free tripwire: eval:bg:gate / eval:bg:periodic --timeout must cover
ceil(shards/jobs) x shard-timeout x 1.05, recomputed from the actual paid
test census every run. Hand-derived numbers go stale every time a paid
file lands — the review split just proved it: periodic's 28800s dropped
BELOW its new 32130s worst case (raised to 32400s here). An undersized
watchdog kills healthy runs and the tail reports never-started.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: preflight ping once in the sharded parent, not per shard
The Anthropic fail-fast ping ran at module load in every paid test file
importing e2e-helpers — ~30 paid claude -p calls (30s timeout each) per
full sharded run for one bit of information. The parent now pings once
before spawning shards and sets EVALS_PREFLIGHT_OK=1; the module-load
path honors the flag. Extracted to test/helpers/anthropic-preflight.ts
(injectable spawn seam) with regression pins in both directions: the
flag must skip, its absence must ping exactly once, dead API must throw.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: split touchfiles into pure data + selection logic + facade
touchfiles.ts listed ITSELF in GLOBAL_TOUCHFILES, so adding one test's
dep entry forced the full ~$38 / 30-45min suite — measured on 21.9% of
recent commits (42/192). The self-reference existed because data and
logic shared a file: any edit COULD be a selection-logic change.
Now: touchfiles-data.ts (the four maps, literals only, zero imports —
the future map-diff target), test-selection.ts (matchGlob/detectBase
Branch/getChangedFiles/selectTests), and touchfiles.ts as a re-export
facade so all ~12 import sites are untouched. GLOBAL_TOUCHFILES drops
the self-ref, adds test-selection.ts (logic stays maximally
conservative), and TEMPORARILY adds touchfiles-data.ts until the
map-diff change lands. New free test pins the literal-only property
(comment-aware state-machine scan with a self-test) and facade export
parity (===), so neither can silently rot.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: free runner — strict output, parallel execution, stable shard indices
Three coupled changes to scripts/test-free-shards.ts:
1. STRICT OUTPUT: runFreeShard streams through the paid runner's
BunTestOutputClassifier — exit 0 without bun's 'Ran N tests across M
files' summary, with (fail) lines, or with a wrong file count is a
FAILURE (anti-truncation backstop at the runner layer), plus an
external wall-clock timeout that SIGKILLs the process group
(timed-out distinct from failed; exit 124 vs 1). Also fixes a latent
shard-bleed: file selectors now use exactTestFileSelectors (relative
paths were substring filters that matched sibling roots).
2. PARALLEL: full-suite mode is one 'bun test --parallel' invocation
(Bun 1.3.13). Measured semantics recorded in the header: per-file
worker isolation, standard summary, and mid-suite process.exit
surfaces as a crashed-worker FAIL with exit 1 — strictly safer than
serial, where the same exit truncates silently. No static weight
lists; --shards M --shard i keeps deterministic hash partitioning for
CI matrices (native --shard rejected: round-robin renumbers when
files land). Spawned shards get throwaway GSTACK_HOME/TMPDIR so
parallel shards can't contend on real state. Per-shard epilogue
prints files/seconds/status every run.
3. Stable indices: assignFilesToShards no longer drops empty shards, so
a shard's index depends only on the file hash and requested count —
an empty CI matrix slot is a fast no-op success, not a renumbering.
package.json 'test' now delegates to the runner (TEST_ROOTS becomes the
single source of truth for roots; slop:diff tail preserved; the runner
inherits the 30s per-test timeout the old glob passed inline).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci: Linux free-test lane — ~400 files get CI coverage for the first time
New required, secretless free-tests job: the canonical runner's single
'bun test --parallel' invocation with strict-output classification on
ubicloud-standard-8. The free suite previously ran on NO Linux CI — only
a curated Windows subset ran anywhere — so every 'tests pass' claim
about main rested on contributors running them locally.
Secretless by design (no API keys; fork PRs finally get real test
signal) and pinned by test/free-tests-workflow-wiring.test.ts: canonical
runner invoked, zero secrets.* references, pull_request never
pull_request_target, and matrix-count/--shards agreement if anyone
switches to the sharded fallback.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: map-diff selection — a touchfiles-data edit runs only what changed
Editing the eval dep-list data no longer forces the full ~$38 /
30-45min suite (measured on 21.9% of recent commits). When
touchfiles-data.ts is in the diff, selection now evaluates the BASE
version (git show -> mkdtemp -> spawnSync bun child printing the four
maps as JSON — sync because e2e-helpers selects at module scope) and
JSON-diffs per key: added entries, edited dep lists, and tier flips are
selected; keys removed from all maps are reported, never silently
dropped; a GLOBAL_TOUCHFILES edit still runs everything.
FAIL-CLOSED with named causes: missing-base-ref, git-show-failed,
import-failed, shape-mismatch each degrade to run-all and print
'selection: global — touchfiles-data changed (<cause>)' (D9 — silently
expensive beats silently wrong, but never silently). eval:select prints
'selected N of M, reason: ...' + removed tests; --base scopes the
map-diff too.
The temporary conservative GLOBAL entry for touchfiles-data.ts is gone —
its changes route through the map-diff. 23 new free tests: pure-core
fixtures, selectTests wiring incl. a poison-injection guard, and a temp
git repo exercising every fail-closed cause end-to-end.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: selection sees uncommitted work; git errors fail closed
getChangedFiles is now the deduped union of committed (base...HEAD),
staged+unstaged (git diff HEAD), and untracked (git status --porcelain
--untracked-files=all) — an agent that edits files and runs evals
BEFORE committing no longer gets the full $38 suite every time because
the committed diff looked empty. Clean tree still returns [] (run-all
by design for main-branch/periodic runs).
Git failures now THROW with the failing command, stderr, and 'set
EVALS_ALL=1 to deliberately run the full suite' — the old return []
silently became run-all, which is silently expensive. 11 new free tests
cover every source, dedupe, quoted paths, and both failure shapes via
an injectable spawn seam.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: revert GSTACK_HOME injection in the free runner — shared mutable state
The first full run under the strict runner surfaced 12 failures with one
root cause: injecting a single throwaway GSTACK_HOME per invocation made
6,900 tests share a MUTABLE scratch home. gstack-config tests wrote keys
into it; relink and update-check tests then read them (e.g. relink saw
skill_prefix left behind by a config test and produced prefixed names).
All 12 pass when run directly.
TMPDIR isolation stays (mkdtemp inside it is still per-call unique).
Tests needing GSTACK_HOME isolation mkdtemp their own per test — the
repo convention — and hermetic-env covers E2E children. The env-dump pin
now asserts GSTACK_HOME passes through UNTOUCHED so the injection can't
come back.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: rebase parity baseline to v1.64.0.0; fix capture-vs-check drift
The parity ratchet had quietly failed for 7 skills — v1.58-v1.64 growth
landed past the v1.57.7.0 anchors and nothing caught it because this
test had no CI lane (verified pre-existing: SKILL.md content is
byte-identical to origin/main). Same rebase protocol as
v1.53->v1.57.7.0; old baseline retained for the audit trail.
Root-caused a second latent bug while rebasing: captureBaseline recorded
SKELETON-ONLY bytes while the checker compares UNION bytes (skeleton +
carved sections/*.md), so a fresh capture read carved skills at ~2x
ratio (ship: 82KB captured vs 183KB checked). captureBaseline now takes
sectionedSkills and records unions for carved skills — capture and check
measure the same thing, so the NEXT rebase can't hit this. Four
CARVE_GUARDS skeleton caps re-ratcheted to current +headroom
(plan-ceo 92K, plan-eng 70K, office-hours 100K, design-consultation
70K), annotated inline.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: package.json version matches VERSION (1.64.0.0)
v1.64.0.0 shipped with VERSION bumped but package.json left at 1.63.0.0
— the 'package.json version matches VERSION file' test fails on
origin/main today. Nothing caught it because that test had no CI lane
until this branch's free-tests job.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: fix variants-retry-after HTTP-date flake (TODOS P2)
toUTCString() truncates to whole seconds, so a +3000ms Retry-After date
could mean an effective wait of ~2001ms — flaking against the 2500ms
assertion floor ~1-2 in 9 runs under suite load. +4000ms puts the
truncation floor at 3001ms with the assertion floor safely below it.
Pulled forward from U4 because the free-tests lane is now a required
check and this flake would randomly block PRs.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: skill-fixture helper — extract SKILL.md sections, don't copy files
extractSkillSections (fence-aware H2 scanner, loud-throw on missing
sections with available-heading list), extractSkillBody (drops the
shared generated preamble), extractSkillHead (frontmatter + first 30
lines, for routing fixtures). Pinned section lists per consumer, and
free-tier real-skill pins so a gen-skill-docs heading rename fails the
FREE suite instead of a paid run. skill-fixture.ts joins
GLOBAL_TOUCHFILES (fail-safe polarity: over-select).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): review E2E fixtures extract sections — 1871 -> 207 lines
CLAUDE.md's extract-don't-copy rule, applied: the three review fixtures
carry only the sections the sql-injection/enum/design-lite prompts and
judges exercise (89% cut). Full-file copies made claude -p read 1871
lines per test — the direct cause of the 1705s worst shard (94.7% of
the 1800s kill).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): retro E2E fixtures extract sections — 1821 -> 757 lines
Keeps every section the retro flow exercises incl. base-branch detect;
drops preamble, Global Retrospective Mode, Compare Mode (58% cut).
retro-base-branch was the single slowest CI test at 224s.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): review-army fixture extracts sections — 1871 -> 650 lines
CS1's set plus Step 1.5 (PLAN COMPLETION AUDIT machinery) and Step 4.5
(army dispatch, quality_score, findings schema) that the 7 army tests
assert on. Pin test guards the three load-bearing strings.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): skillify fixtures via extractSkillBody — 63-83% smaller
Tests follow all 11 skillify steps, so the whole body stays; only the
shared generated preamble drops (skillify 1239->453, scrape 958->167).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): context-skills fixtures via extractSkillBody — 74-82% smaller
context-save 1037->267 lines, context-restore 952->168; the 8 tests
exercise full save/restore/list flows so the body stays, preamble drops.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): opus-47 discovery fixtures via extractSkillHead — ~95% smaller
Routing/fanout tests only read frontmatter + opening lines of the 14
installed skills (review 1871->54, office-hours 1706->80).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): codex runner gains sections option — review variant 88% smaller
runCodexSkill/installSkillToTempHome accept sections?: string[] routed
through extractSkillSections; codex-review-findings wired (1465->181
lines). codex-discover-skill deliberately keeps the FULL copy — its
stderr assertions validate that the real generated artifact loads.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test(evals): routing fixture installs skill HEADS, not ~18 full SKILL.md
Routing reads frontmatter only; extractSkillHead per skill (root
611->48, ship 1435->54 lines). This was the single worst fixture bloat
site: one fixture dir holding ~18 full skills.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: parent-side shard skipping — a one-test diff runs 3 of 44 shards
The sharded runner spawned every shard regardless of diff; only the
child self-skipped, so a typical single-skill change still paid 44 Bun
boots + container-equivalent setup for shards with zero selected tests.
The parent now computes selection once (mirroring e2e-helpers exactly:
EVALS_ALL -> run-all, empty union -> run-all, git errors propagate the
fail-closed throw) and drops shards where no selected test name maps in.
Mapping = quoted E2E map keys in the file's source UNION keys whose dep
list registers the file (constructed-name families need the second
direction). FAIL-OPEN everywhere it matters: run-all, non-skill-e2e
files, unreadable source, zero mapped names all keep the shard — the
child filter stays authoritative, so a parent bug can only run extra.
New taxonomy status skipped-by-diff (never conflated with
never-started); selection banner prints once; --list is selection-aware.
C6 lands in the same commit: a HARD tier-alignment test — every paid
skill-e2e file must be parent-mappable or provably fail-open-safe.
Note: this change-set's 14 dep-list registrations in touchfiles-data.ts
rode along in f945c841 (concurrent-agent staging); they belong to this
change logically.
Demo: selection of one test -> 'running 3 of 44 shards, 41
skipped-by-diff'. 13 new $0 tests via injected seams.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: fix context-save-list test that was 0-for-26 ($5.28, zero passes)
Disposition for the eval store's only permanently-red test. Root cause:
the hide-other-branches assertions scanned the FULL output surface
(incl. bash tool_results), so any agent that ran ls on the checkpoints
dir — the natural first step of a list flow — surfaced all three seeded
filenames and failed, even when its user-facing listing filtered
correctly. The test punished the agent for looking at the directory.
The hide-assertions now scan the agent's FINAL TEXT (the listing the
user sees); showsMain keeps the broad surface for its documented reason.
Validated live in the final gate run rather than quarantined: the test
guards real behavior (branch filtering) and the assertion was the bug.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: wire 12 orphaned test files into the free suite (D3a)
ios-qa/daemon/test (10 files), ios-qa/scripts/gen-accessors.test.ts, and
browser-skills/hackernews-frontpage/script.test.ts ran under NO script
or CI — written coverage catching nothing. All 174 tests green on
arrival (4.6s), zero quarantines needed. TODOS P2 closed: main wired
design/test in v1.64, the variants-retry-after flake it named is fixed
on this branch, and this commit lands the remaining orphans.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: supabase-provision runs in-process — 16.5s -> 0.45s
bin/gstack-gbrain-supabase-provision (482-line bash) becomes a 26-line
bun-shebang entry over a new importable lib/gbrain-supabase-provision.ts
with an injected-deps seam (fetch/env/stdout/sleep — D7: args, never
env-mutation-before-import). The 33 spawn-per-test cases run in-process
against the same Bun.serve mocks; exactly one spawn smoke keeps the
shebang/CLI/receipt contract covered.
Byte-compat proven by a 25-case differential harness (old bash bin from
git vs new, same mock): stdout, stderr, exit codes identical across all
subcommands, JSON/plain modes, and error paths. Egress receipts stay
per-attempt, receipt-before-send, fail-closed (scanner updated:
SHELL_SINKS -> MODULE_SINKS). No-op sleep injection makes retry/backoff
paths instant.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: kill the 3,372-line zombie monolith; revive 4 never-run tests
test/skill-e2e.test.ts survived the v1.56 split as a zombie: the paid
glob needs the skill-e2e-* hyphen, so with EVALS=1 NOTHING has executed
it for ~8 releases — and it held the ONLY implementations of four
map-registered tests: review-coverage-audit (gate), plan-eng-coverage-
audit (gate), ship-triage (gate), ship-idempotency (periodic). Three
gate tests silently never ran — the exact 0%-execution class this
branch exists to kill.
Rehomed into test/skill-e2e-coverage-audit.test.ts, -triage.test.ts, and
-ship-idempotency-sdk.test.ts with bodies byte-identical modulo collector
wiring and fixture extraction (drift observed in the skills since v1.56
is DOCUMENTED in each header, not fixed — their first paid run in 8
releases must attribute failures to drift, not to this move). All 24
other monolith names were true duplicates of the split files — dropped
with the monolith. Matrix rows added to both eval workflows; the paid
glob's zombie-exclusion is now a commented regression pin.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: fix two parallelism-exposed flakes (probe re-run, live-tree census)
Both pass solo and on main but flaked under the parallel runner:
1. gstack-brain-context-load probed 'gbrain --version' PER QUERY with a
500ms budget — a cold probe on a saturated box timed out (observed
505ms), branding gbrain 'missing' for one query while siblings
passed. The probe is now memoized (availability can't change
mid-invocation) with a generous one-time 5s budget; query calls keep
the tight timeout.
2. skill-size-budget's catalog estimate read the LIVE tree, so a
concurrent worker's transient skill-shaped scratch dirs exactly
doubled it (8356 vs 4177). The ratchet now counts git-TRACKED skills
only — the catalog that ships, immune to sibling workers.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: demote 4 expensive posture tests to periodic (D2a)
design-consultation-research ($0.91/304s) and -preview ($0.89/481s) —
the two most expensive gate tests — plus office-hours-forcing-energy
(LLM-judge posture score; its sibling was already demoted) and
cso-full-audit (250s/$0.57; the targeted cso tests stay gate). Saves
~$8-12 and 10-15 min per gate run. The plan-*-finding-floor tests stay
gate deliberately: cheap insurance on the most-edited skill surface.
Housing files have no whole-file self-gates, so the runtime E2E_TIERS
filter handles both tiers; tier-alignment tripwire green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: judge default Sonnet -> Haiku 4.5 (D1a)
The 25 doc-quality judges are rubric-scoring calls — a duty Haiku is
already proven at in this repo (pty hung/working classifier,
first-task-scaffold, hermetic-canary). Tests needing a stronger judge
pass a model explicitly. Note: eval-store judge costs were hardcoded
synthetic (0.02), so no baseline distortion. Re-baselined by the
periodic run in this branch's final verification.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: SDK runner default Opus -> Sonnet (D1a)
agent-sdk-runner defaulted to Opus 4.7 while session-runner (the claude
-p path) defaulted to Sonnet — an inconsistency between the two runners,
not a decision anyone made. Unpinned tests were implicitly asserting the
expensive model. The 30+ tests that genuinely need Opus already pin it
via opts.model. Re-baselined by the periodic run in this branch's final
verification; regressors get explicit Opus pins.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: CLAUDE.md tells the truth about the free suite; make-pdf gate is macOS-only
The '<2s' claim was off by two orders of magnitude (measured 454s serial
at v1.63; ~90-100s now under the parallel runner), and the bare
'bun test' guidance walked the whole repo, loading paid eval files and
missing the strict classifier. Commands now say 'bun run test' with real
numbers, document the strict-output invariant, the EVALS_JOBS /
EVALS_CONCURRENCY split, the computed detach-timeout floor, and the
required free-tests lane. make-pdf-gate drops its Linux leg (redundant
with the free lane running make-pdf tests on every PR); macOS rendering
coverage stays.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: catalog ratchet reads committed content; SDK unit pins follow D1a default
Two follow-ups from the verification runs:
1. skill-size-budget's catalog estimate still flaked under --parallel
(8356, then 8041, vs 4177 solo) even after filtering to tracked
skills: sibling workers REGENERATE real SKILL.md files mid-run, so
any live-tree read is a moving target. The ratchet now reads each
tracked skill's frontmatter from git show HEAD: — the catalog that
ships — which no concurrent worker can perturb.
2. agent-sdk-runner unit pins asserted the old Opus default through the
default-flow fixtures; flipped to the Sonnet default (the explicit-
override pass-through pins keep Opus — that path is unchanged).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(browse): SIGKILL abandoned Chromium on close-race timeout (suite wedge)
close()'s launched-mode path raced browser.close() against 5s and on
timeout ABANDONED the child: this.browser nulled, process handle lost,
Chromium alive holding keep-alive connections into test servers whose
stop() then waits forever. Reproduced twice as an intermittent (~50%)
whole-suite wedge — a 44min 0.1%-CPU hang pinned by a leaked LISTEN
socket, and a 400s hang with commands.test.ts teardown in flight.
The child handle is now captured BEFORE the race and SIGKILLed on
race-timeout (launched mode only; headed keeps context.close). Race
timers are unref'd so a successful close stops pinning the caller's
event loop for the window. The four browse test servers force-close
keep-alives (stop(true)) as belt-and-braces.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: delete two dead-architecture security contract tests
browse/test/security-source-contracts.test.ts and sidebar-security.test.ts
read browse/src/sidebar-agent.ts at module scope — a file deleted (on main
too) when the sidebar chat-queue path was ripped in favor of the terminal
PTY. Both files have errored on load ever since: the old truncating suite
never surfaced it, and no CI lane ran them. Their subjects (queue-spawn
canary injection, preSpawnSecurityCheck, queued args, chat system prompt)
no longer exist; server.ts retains processAgentEvent only in a comment.
Live security coverage continues in security.test.ts (canary/verdict),
content-security.test.ts (L1-L3), server-sanitize-surrogates.test.ts,
and the security-bench suite. If the terminal-agent path should inherit
any of the deleted contracts, that is a separately scoped piece of work
against the component that actually exists.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(redact): calibrate placeholder recognition for code and doc shapes
Three pushed-secret false positives blocked this branch's push; each is
now recognized as a placeholder in the url_with_password/basic_auth_url
validators, with real passwords still blocking (all pinned):
- ${camelCase} JS template interpolations (the old check only skipped
uppercase env-style ${DB_PASS}, so the supabase-provision bash->TS
port's `postgresql://${dbUser}:${dbPass}@...` flagged as two
pushed secrets).
- The literal PASSWORD/pass placeholder in URL-format doc comments.
- The provision lib's doc comments now use <PASSWORD>/PASSWORD forms.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: opt-in gate for live-playwright ML tests; ios-qa build hygiene
security-live-playwright's L4 tests dlopen onnxruntime inside a bun
--parallel worker whenever the dev box has a warm model cache — the
source of the intermittent 'panic: Segmentation fault' + crashed-worker
retries (and likely the residual run wedges). Same SECURITY_BENCH=1
opt-in as security-bench.test.ts; the L1-L3 tests in the file still run
everywhere.
Also: gitignore the ios-qa gen-accessors-tool Swift .build/ output (a
side-effect of running its tests that kept polluting git status) and
commit its Package.resolved so tool builds resolve reproducibly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: free runner output contract — name the failure, quiet the noise
Diagnosing a red run used to mean re-running with output captured to a
file and grepping past ~1000 lines of tab-close spam and ASCII art —
several runs today ended with no way to even NAME the failing test, and
a wall-timeout kill said nothing about which file wedged.
New contract: the full child stream ALWAYS lands in a per-run log file
(path printed up front); the console shows only runner lines, (fail)
results, crash markers, and the terminal summary (--verbose restores
the firehose; the strict classifier consumes the full stream in every
mode). After every run a stable epilogue names the outcome:
[test:free] FAIL — k failing test(s) in j file(s), c crashed
worker(s). Full log: <path>
✗ <file> — <test name>
⚠ crashed+retried: <file>
⏱ in flight at kill: <files> (timeout only — the wedge suspects)
Attribution rides bun --parallel's per-file output grouping
(ANSI-stripped — color codes defeated a plain grep today). 12 new pins:
epilogue formats, crash surfacing, quiet/verbose console policy, log
completeness, in-flight-at-kill on a real hang.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: quarantine 5 pre-existing env failures individually (receipts in-file)
Three snapshot tests (stale-ref error, snapshot -D diff, annotation
cleanup) and two extension-sender-auth behavioral tests fail identically
on origin/main v1.64.1.0, solo, on dev machines — verified per the blame
protocol. Main's CI lane skip-lists both FILES wholesale; quarantining
only the five failing tests keeps the other 60 guarding. Each carries
the un-skip condition.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: stealth-webdriver launch gets parallel-load headroom (120s)
Playwright's default 30s launch timeout dies under the full-suite
--parallel run when ~400 workers contend for Chromium launches — bun
reports the hook death as an '(unnamed)' 30006ms failure (named on
sight by the new runner epilogue). Both launch sites get explicit 120s
timeouts; the runner's external wall-clock still bounds the ceiling.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: free-runner wall timeout 15min -> 6min (faster wedge diagnosis)
The suite completes in ~100-160s; a wedge used to mean 15 minutes of
silence before the kill-and-name epilogue fired. 6min keeps ~3.5x
headroom over the slowest observed clean run while naming wedge
suspects in minutes. --wall-timeout <secs> overrides per run.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: run worker-hostile files in a serial child (first entry: security-live-playwright)
The residual full-suite wedge, named by the new epilogue: Bun 1.3.13
segfaults running browse/test/security-live-playwright.test.ts in a
--parallel worker ('panic: Segmentation fault ... a bug in Bun'), and
the crashed-worker retry then wedges the whole invocation past the wall
clock. The file passes serially.
New WORKER_HOSTILE placement list: full-suite mode excludes listed files
from the parallel invocation and runs them in their own strict-classified
serial child afterward — execution placement, not a skip; each entry
carries its reason and removal condition.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: gate compare-board's file-level hooks too — the intermittent staller
Skipped describes do NOT skip file-level hooks: the quarantined
compare-board file still ran its top-level beforeAll (PNG fixtures +
Bun.serve + a BrowserManager launch — exactly the 'needs a
display-shaped env' code) on every run, and under parallel load that
setup wedges. Caught red-handed by the runner's in-flight-at-kill
epilogue: '⏱ in flight at kill: browse/test/compare-board.test.ts'.
This was the suite's intermittent staller. Hooks now honor the same
GSTACK_COMPARE_BOARD_TESTS gate; the gated file drops from 3.7s of live
setup to 0.4s of pure skips.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: full suite runs as N shard processes; scrub spec-sync child env
Two fixes from the wedge-hunt endgame:
1. Full-suite mode switches from one 'bun test --parallel' invocation to
N concurrent shard PROCESSES, serial within each (the paid runner's
proven model; N = min(6, cpus-2)). The single-invocation strategy hit
three distinct Bun 1.3.13 worker pathologies in one day — a segfault
whose crashed-worker retry wedged the run, a quarantined file's
still-running file-level hooks stalling a worker, and spawn-heavy
files hanging under load — and each one stalled the WHOLE invocation.
Process shards isolate any wedge to its own shard. First full run
under this model: no wedge, six epilogues, one real failure named.
WORKER_HOSTILE stays as the paper trail; --parallel remains available
per-shard for a future Bun.
2. That one real failure: spec-template-sync regenerates SKILL.md via a
child that inherited the shard process's env — an earlier test's
GSTACK_*/GBRAIN_* mutations changed generator output (failed in-suite,
passed solo on an identical tree). The child now gets a scrubbed env:
generator output must be a function of the templates, not of whichever
test ran before.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: tree-mutating tests run after the parallel shards; scrub relink env
The flake family's root cause, finally: five test files REGENERATE
shared repo artifacts in place (catalog-mode-full rewrites every
SKILL.md in full-catalog mode; spec-sync and idempotency regenerate all
skills; gen-skill-docs and skill-validation rewrite .agents/). Any
concurrent shard reading those files sees a moving target — this one
family produced the exactly-doubled catalog estimate, the golden-file
drift, and the spec-sync mismatch chased earlier today. Full-suite mode
now runs TREE_MUTATING files in one serial shard AFTER the parallel
shards complete; CI's matrix is unaffected (per-runner checkouts).
Also: relink's run() helper spread process.env into its children, so a
sibling file's leaked GSTACK_HOME made the 'fresh install' test see a
neighbor's skill_prefix. GSTACK_HOME is now dropped unless the test
passes it explicitly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: gbrain-detection-override joins TREE_MUTATING (mutator #6)
It regenerates SKILL.md in place with --respect-detection (the gbrain
variant adds ~1-3KB per carved skeleton) and git-restores afterward —
its own header documents the approach. During that window the parity
suite in a concurrent shard read inflated skeletons and failed 4 caps.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: tree-ratchet readers join the serial phase (quiet tree by construction)
Two consecutive runs failed the parity caps with byte-identical inflated
skeletons (+~2KB gbrain-variant blocks) while the tree was clean before
and after — some concurrent regen window keeps escaping the mutator
census. Rather than hunt every present and future mutator, the tests
that MEASURE the shared tree (parity caps, size budgets, carve guards)
now run in the serial phase after the parallel shards: a quiet tree by
construction, immune to any regen we haven't found.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* evals: judge default back to Sonnet — Haiku regressed the rubric family (A/B receipts)
The partial-diff rehearsal was the Haiku judge default's first live run
and it failed all three selected doc-rubric judges. Controlled A/B on
the identical health-rubric prompt: Haiku 2/2/2 vs Sonnet 4/3/4, both
with coherent reasoning — Haiku is simply a harsher grader on
long-document rubrics, and every >=4 threshold in skill-llm-eval was
calibrated against months of Sonnet baselines. Per D1a's
pin-on-regressors protocol the default reverts; a new
GSTACK_EVAL_MODEL_JUDGE override makes future recalibration a one-var
experiment. Haiku keeps the classifier-grade duties (pty hung/working,
warmup, distill via lib/eval-model.ts) and D1a's capture->Sonnet stands.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(test-runner): per-origin classifier buffers — interleaved pipes can't shear lines
stdout and stderr are independent pipes; a chunk from one can arrive
between two halves of a line from the other. The single shared
pending-buffer glued those fragments into garbled lines: a sheared
(fail) line went uncounted (defeating the exit-0-with-failures
backstop) and a sheared terminal summary read as truncation.
Counters stay shared; line assembly is now per-stream, and both
runners tag the stream origin. Also drops the dead ChildProcess
type import left by the killProcessGroup move.
* fix(test-runner): real carve-guard keys in TREE_MUTATING; census pins; size-scaled wall deadlines
TREE_MUTATING listed 'test/carve-guard-checks.test.ts' — a file that
has never existed (the real ratchet readers are
carve-guard-completeness and carve-section-ordering), so the intended
serialization was silently absent. New census pin tests fail on any
key that doesn't name a real free test file, and on a TEST_ROOTS
entry that stops contributing files. Full-suite wall deadlines now
scale with shard size (max(6min, files x 5s)) so a jobs=1 machine or
the ~130-file Windows shards can't false-timeout a healthy run;
explicit --wall-timeout disables scaling. Stale --parallel wording in
the dry-run message, jsdoc, and the TREE_MUTATING ordering comment
corrected to the shipped process-shard model.
* fix(evals): selection under-selection fixes — duplicate keys, self-paths, quotePath
Three under-selection holes: (1) duplicate E2E_TOUCHFILES keys
(ship-plan-completion/-verification) — JS keeps the LAST duplicate, so
the earlier dep lists were dead; pair deleted and a duplicate-key scan
added to the literal-only tripwire. (2) The five rehomed e2e files
didn't list themselves in their own dep lists, so editing the test
never selected it. (3) git C-escapes non-ASCII paths without
core.quotePath=false, so an accented filename matched no glob and
deselected its tests. Also updates the stale --retry cost comment.
* fix(evals): destructive-actions guard actually inspects Bash commands
The rehomed guard filtered on typeof input === 'string', but
session-runner records tool inputs as objects ({command} for Bash) —
the filter matched nothing and the assertion could never fail, even
against a real 'git push'. Now extracts the command from the object
shape, same as the usedGitDiff check above it.
* fix(redact): interpolation allowance can't swallow a real $word password
The placeholder calibration used optional braces on both sides, which
also suppressed bare $lowercase — a real password starting with '$'
would have passed the HIGH gate. Interpolation now means ${identifier}
(braced, any case) or bare $UPPER_SNAKE only; both connection-string
patterns share one validator so they can't drift. Pins added for the
bare-$word block, $UPPER allowance, and mismatched-brace block.
* fix(gbrain): wait --timeout validates up front instead of polling forever on NaN
Number('abc') is NaN, NaN comparisons are always false, and the
poll loop never hit its deadline — an infinite 5s loop where the bash
predecessor errored immediately. die(2) at parse time, with a test.
* ci: least-privilege tokens on the two lanes that execute PR-controlled code
free-tests runs PR code (install lifecycle scripts + the suite) with
whatever the repo-default GITHUB_TOKEN grant is, persisted into
.git/config by checkout. Now: permissions contents:read,
persist-credentials false, pinned by the wiring test. actionlint gets
the same treatment plus a digest pin on the third-party Docker Hub
image (a tag is repointable with no GitHub-side audit trail, and the
image sees the mounted checkout). restore-keys added to both caches so
a lockfile bump warms from the previous cache; stale --parallel header
wording corrected.
* test(browse): unit coverage for the close() SIGKILL fallback
The wedge fix (capture the Chromium child before the close race,
SIGKILL on timeout) shipped without a test of the branch it added —
the coverage audit flagged it as the diff's one regression-gap. The
5s race window becomes an injectable closeRaceMs field, and four unit
tests pin: SIGKILL on hang, no SIGKILL on clean close, no SIGKILL on
an already-exited child, SIGKILL on a rejecting close.
* docs: CLAUDE.md describes the shipped shard-process model, not the abandoned --parallel probe
* fix(test-runner): cancellation terminates the run; win32 kills the whole tree
Installing SIGINT/SIGTERM forwarders suppresses Node's default
terminate-on-signal, so a cancelled run killed the current child and
kept LAUNCHING shards — observed as paid runs continuing to burn API
spend after Ctrl-C (codex adversarial, repro'd ALIVE_AFTER_SIGTERM).
The first signal now also schedules the parent's own exit after the
children's SIGKILL grace, and both shard pools consult
isTerminationRequested() before taking new work. On win32,
killProcessGroup uses taskkill /T /F — detached:true creates no
killable group there, and a bare child.kill orphaned every grandchild
(ports, locks, and the inherited pipes that kept close from firing).
Also: the tree-mutating serial shard prints dirty generated artifacts
when it dies mid-regeneration, and --shard CI-matrix mode gets the
same size-scaled wall deadline as full-suite mode.
* fix(evals): preflight fails fast on spawn error, timeout, and exit 127
The ping only grepped stdout for two connection strings — a missing
claude binary, a 30s timeout kill, or command-not-found all returned
'ok', and the fleet then burned ~30 shard timeouts discovering the
outage one child at a time. Cross-model finding (testing specialist +
codex adversarial). Other non-zero exits stay deliberately fail-open:
a flaky preflight must not block a runnable suite; pinned both ways.
* fix(redact): lowercase 'password'/'pass' at the URL-password position blocks
The case-insensitive placeholder words waved postgres://admin:password@host
through the HIGH gate as a doc placeholder (codex adversarial,
verified zero findings pre-fix). URL-password position is now stricter
than generic placeholder detection: ALL-CAPS doc convention
(USER:PASSWORD), ${identifier} interpolations, bare $UPPER_SNAKE, and
structural shapes (<your-password>) suppress; lowercase dictionary
words block. Pinned in both directions.
* fix(gbrain): DSNs percent-encode the password; body reads retry; stdout drains
Three codex-adversarial findings in the provision port: (1) raw DB_PASS
interpolation — a reserved character (/ # ? % @) restructured the URI,
provisioning succeeded, and every consumer then failed to parse the DSN
(unusable billable orphan); now encodeURIComponent, round-trip pinned.
(2) await res.text() sat outside the transport try — a server that sent
headers then reset the stream was an uncaught exit 1 instead of a
retry-then-exit-8. (3) The bin entrypoint called process.exit() after
unawaited stdout writes, truncating piped JSON; exitCode lets writes
drain.
* fix(evals): selection-path helpers join GLOBAL_TOUCHFILES; base-branch keys self-register
The three-file split moved test-selection.ts into the globals but
dropped the facade — an edit to test/helpers/touchfiles.ts (executable
selection-path code imported by every consumer) selected ZERO paid
tests, the exact invisible-non-execution class this branch exists to
kill (claude adversarial, finding 1). e2e-helpers.ts (the harness every
paid test imports) and paid-test-set.ts (paid-vs-free classification)
had the same gap. The review/ship base-branch keys also register
test/skill-e2e-review-attribution.test.ts so editing those tests
selects them.
* ci(free-tests): PR-number concurrency, failure-log artifact, main-push runs
Three red-team/adversarial findings on the new required lane:
(1) concurrency keyed on bare head_ref — two forks with the same
branch name shared one group, so a push to fork B cancelled fork A's
in-flight REQUIRED check (merge-pipeline DoS with no code fault); key
on the PR number. (2) The runner's full logs die with the runner in
os.tmpdir() — a red check named WHICH test failed but never why;
upload the shard logs as an artifact on failure. (3) PR-only trigger
meant two individually-green PRs could merge into a red main with
nothing running the suite there; add push: branches: [main].
* ci: PR-number concurrency keying on the eval and Windows lanes too
Same fork-branch-name collision as free-tests.yml: bare head_ref
carries no owner prefix, so same-name branches from different forks
shared a cancel-in-progress group.
* test(evals): retro E2E passes require the report on disk
Both retro tests passed with zero work product: error_max_turns
counted as success and the content assertion was guarded by
fs.existsSync — a run that burned 30 turns and wrote nothing recorded
green (red team). The report is now load-bearing for pass/fail.
* chore: bump version and changelog (v1.66.0.0)
Test/evals/CI speedup pass: release summary + itemized changes in
CHANGELOG.md; TODOS.md marks the free-suite exit-code P1 complete and
files the review-army follow-ups.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(redact): fully-braced ${...} interpolations are code, whatever they contain
The identifier-only braced form flagged the DSN builder's own
${encodeURIComponent(dbPass)} call site as a pushed secret — a scan
that cries wolf on the fix for the previous finding. Any ${...}
spanning the whole password segment is template code; bare $word
stays uppercase-only so $hunter2 still blocks. The mismatched-brace
negative fixture assembles at runtime so this file's own pushed bytes
carry no blockable URL shape.
* test(gbrain): assemble the pooler expected-URL from parts (scan-clean pushed bytes)
* docs: sync docs for v1.66.0.0 (test/evals/CI speedup)
CONTRIBUTING.md, AGENTS.md, and ARCHITECTURE.md still taught bare
`bun test` for the suite; the shipped runner deprecates it (walks the
whole repo, loads paid eval files, misses the strict classifier). All
suite-level references now say `bun run test`, the Tier 1 section
describes the strict shard runner (~90-100s, --verbose, --wall-timeout),
the sharded paid-runner paragraph documents diff-based shard skipping
and the EVALS_JOBS / EVALS_CONCURRENCY split, the Tier 3 row points at
the actual judge-only invocation, and GSTACK_EVAL_MODEL_JUDGE is
documented at the judge it overrides.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* ci(free-tests): restore the PR-number concurrency + failure-log artifact; truth-fix stale comments
The workspace-revert incident that hit CHANGELOG/TODOS mid-ship also
caught free-tests.yml between edits: commit 8d6c2ff8's message claims
PR-number concurrency + artifact upload + main-push runs, but only the
push trigger survived to the commit (caught by the /document-release
doc-vs-code audit). Both re-applied. Also: eval-model.ts header said
capture defaults to Opus (it's Sonnet per D1a), paid-shards' header
pinned a stale 44/63 shard census, and two CHANGELOG phrases
over-claimed ('six' -> 'up to six' shard processes; retry-1 scoped to
retry-bearing paid paths).
* ci: setup-buildx before every cache-exporting image build
First live run of the cache trio failed at flag-parse time: the
default buildx `docker` driver hard-errors on cache-to registry
export ('Cache export is not supported for the docker driver'), which
failed build-image on PR #2593 and skipped the entire gate eval
matrix behind it. docker/setup-buildx-action creates the
docker-container builder that supports registry cache export; all
three build sites (evals, evals-periodic, ci-image) get it.
* fix(browse): Xvfb identity is argv[0]'s basename, not a cmdline substring
First Linux CI run: isOurXvfb identified the TEST RUNNER as our Xvfb —
the suite's own argv contains 'xvfb.test.ts', the substring match over
the whole cmdline passed, and the start-time check matched because the
pid was real. Any process whose ARGUMENTS mention xvfb (a runner, an
editor) was killable — the sibling-kill class the identity check
exists to prevent. Identity now rests on argv[0]'s basename ('Xvfb'),
with a sh-$0 regression pin. isDisplayFree falls back to the X
socket/lock files when xdpyinfo isn't installed (x11-utils is absent
on some images that ship Xvfb).
* fix(test-runner): strip GHA ::group:: wrappers before file attribution
On GitHub Actions bun wraps each file's log section in ::group::. The
un-stripped header failed FILE_HEADER_RE, failures attributed to the
PREVIOUS file, and the terminal recap's re-printed (fail) lines landed
under a phantom second file — the first Linux run reported 5 real
failures as 10 across 2 files (one of them innocent). Strip the prefix
before matching; the existing file+test dedupe then absorbs the recap.
* test: first-Linux-run environment fixes — bun-only PATH shim, claude gate, darwin-scoped pdf gates
Three environmental assumptions the Linux lane exposed:
(1) gbrain-detect's deterministic SAFE_PATH lacked the bun runtime, so
every env-shebang spawn exited 127 on CI; a scratch dir holding ONLY a
bun symlink joins the PATH (appending bun's real dir would leak its
siblings — dev boxes keep gbrain there too).
(2) host-config's 'detect finds claude' assumed a claude binary; the
secretless lane deliberately has none — gated on Bun.which.
(3) The four make-pdf render gates hard-required prerequisites on ANY
CI, but the make-pdf gate workflow is macOS-only by decision and the
Linux lane doesn't build dist/pdf — hard-require scoped to darwin.
* ci(free-tests): run the suite under xvfb-run
Headed-browser tests (handoff, extension sidepanel DOM) need a real
DISPLAY; the first Linux run died on Playwright's 'headed browser
without an XServer' banner. xvfb-run -a provides the display; x11-utils
ships xdpyinfo for display probing.
* fix(test-runner): bun's headerless failure recap can't invent a phantom failing file
Round-3 CI showed the remaining half of the recap bug: bun prints
'N tests failed:' then re-prints every (fail) line with NO file
headers, so they attributed to the stale currentFile — an innocent
file (test/uninstall.test.ts) was charged with another file's 5
failures. The recap marker now ends attribution (currentFile=null,
chunk closed) and recap re-prints of already-recorded test names
dedupe; a recap-only failure the main run never attributed still
records, unattributed, as belt and braces.
* test(browse): sidepanel DOM suite launches with --no-sandbox on CI + console capture
The suite's raw chromium.launch had no --no-sandbox — every browse
test that goes through gstack's launcher (which always passes it)
survived the Linux lane, while this file's sandboxed renderer died on
first navigation: waitForFunction hung to the 15s test timeout, then
every newContext failed with Target.createBrowserContext. Also wires
pageerror/console-error capture at all six pages so a page-side
failure reads as itself in CI logs instead of a bare timeout.
* test(browse): delete the sidepanel security-DOM suite — it tests UI removed in v1.14
Another member of the never-ran class: the file skipped everywhere
(Playwright chromium absent locally, no Linux CI until this branch),
so it rotted invisibly through THREE contract changes — the v1.63
/extension-token bootstrap, the endpoint growth (/memory,
/pty-session, /sse-session), and finally the v1.14 sidebar-REPL
rewrite that removed the security shield/banner UI it asserts on
(#security-shield survives in sidepanel.html as a dead hidden stub
with no JS driver; sidepanel.js:87 and :1317 document the removal).
The Linux lane executed it for the first time and it can never pass:
the behavior is gone. The L1-L3 security filters it name-checked stay
covered by the ~83 unit/behavioral security tests. The free-tests
lane also vendors xterm assets (bun run vendor:xterm) so the
sidepanel terminal scripts load for any future DOM coverage.
* ci(evals): per-row retry override — two receipted rows keep the third attempt
Three PR rounds of receipts: pty-plan-smoke failed attempt 2 in two
consecutive rounds with ROTATING members (plan-design-review, then
plan-eng-review) and e2e-workflow's document-release timed out on
attempt 2 in round 4 — while both families pass on branches still
running three attempts, and every other row stayed green at --retry 1
across all rounds. Matrix rows gain an optional retries field
(default 1); only these two rows set 2, keeping the measured
retry-amplification win everywhere else.
* test(windows): curate the seven POSIX-bound files the expanded lane surfaced; fix flag-utils path embedding
First full run of the expanded Windows lane (13 -> ~258 files, PR #2593
run 31918591602) failed in exactly 8 files. One was a real test bug,
fixed: design-flag-utils embedded a raw Windows ROOT into a bun -e
string where backslashes act as escapes (D:\a\gstack imported as
D:agstack) — forward slashes work on every platform. The other seven
are POSIX-bound in ways the content patterns cannot see (sed/ln/bash
ARE their subject, a shebang shim arrives via variable, wall-clock
retry bounds on the slowest runner) — each gets a receipted
KNOWN_WINDOWS_INCOMPATIBLE entry, and the census pin now covers that
list so a renamed file fails the suite instead of silently keeping a
stale exclusion.
* test(windows): curate skill-census + browser-manager-unit; surface unhandled errors in the epilogue
Round-2 Windows census (zero failing TESTS — the first curation wave
held): shard 1 failed on an unhandled module-load throw in
skill-census (the skills-tree symlink layout needs Developer Mode CI
runners lack) and shard 2 wedged to its wall deadline inside
browser-manager-unit — both get receipted exclusions; macOS + Linux
lanes keep covering the files. The unhandled-error class also exposed
an epilogue gap: it fails the shard via the strict classifier but
produces no (fail) lines, so the epilogue read 'FAIL — 0 failing
test(s)' with no culprit. The reporter now attributes each
'# Unhandled error between tests' marker to its chunk and the FAIL
line carries the count.
* docs: file the two Windows-lane follow-ups (browser-manager wedge, skill-census symlinks)
* test(windows): round-3 curation — seven files the round-2 wedge had been truncating
The browser-manager-unit wedge was cutting shard 2 short, so each
Windows round revealed the next segment of never-run files. With the
wedge excluded, shard 2 completes (50s) and shows its real failures:
seven more POSIX-environment files (PID/cmdline identity probing,
bash scripts as the subject under test, env-scrubbed bun spawns).
Shards 1 and 3 (including all tree-mutators) now PASS on
windows-latest — this should be the fixed point: ~234 files of real
Windows coverage vs the 13 hand-picked before.
* test(windows): round-4 curation (spawnSkill env, symlink fixtures) + shard-log artifact
Shard 2 ran all 132 files with zero (fail) lines yet bun exited 1 —
unhandled errors in a shape neither counter names, and the Windows
lane had no log artifact to attribute them. Statically attributed and
excluded: browser-skill-commands (spawnSkill spawns bun with a
constructed env; resolution fails under Windows spawn) and
security-audit-r2 (evil-link symlink fixtures need Developer Mode).
The lane now uploads its shard logs on failure like free-tests.yml,
with os.tmpdir() pointed at runner.temp so the glob can find them.
* evals: Opus pin on the spec AUQ-matrix entry — D1a regressor, receipts in-file
The periodic re-baseline for the capture default (Opus -> Sonnet)
found exactly one regressor across the seven-entry AUQ behavioral
matrix: spec failed twice under Sonnet ('never reached a question in
budget', 242s) while its six siblings passed; the controlled Opus
re-run passed cleanly (7/7 format, substance 5, 160s), and a second
run through the new per-entry model plumbing confirms. MatrixSkill
gains an optional model field wired into captureFirstAuq; only spec
sets it. TODOS gains the re-baseline receipts for the never-baselined
periodic tail (three setup-gbrain files + ship-idempotency, all
local-only).
* test(evals): scope-gate assertion carries its evidence tail; file the detector-flake TODO
The plan-design-review member fails ONLY scopeGateQuestionObserved
intermittently on unchanged code (PR #2593: red rounds 3/11 + rerun,
green rounds 5/6 — every attempt terminal, no plan-mode leak), and a
bare Expected-true/Received-false is undiagnosable from CI logs. The
check now throws with the last-2KB visible evidence, so the next
failure distinguishes a detector-sensitivity miss from a real silent
bypass. TODO filed with the full receipt trail.
* test(evals): review-dashboard-via budget 300s -> 360s — third ratchet of the same contention story
PR #2472 documented the 180s deterministic 0-turn startup timeouts and
ratcheted to 300s; PR #2593 hit 302s timeouts on attempt 2 in two
consecutive runs while five sibling rounds passed — marginal at 300s
under 40-way in-shard concurrency. Same headroom its contention-class
sibling (retro-base-branch) carries; outer bun timeout rises to 480s.
* test(evals): document-release budget 180s -> 300s — same contention ratchet, receipts in-file
Timed out at exactly 180s on its final attempt twice on PR #2593
(rounds 4 and 13) while passing four other rounds — a 30-turn
multi-step doc workflow is marginal at 180s under 40-way in-shard CI
concurrency. Same story and same fix as review-dashboard-via and
retro-base-branch; outer bun timeout rises to 360s.
* docs: file the systemic in-shard-concurrency follow-up behind the timeout-flake family
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
