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(ship): name the /document-release subagent at every Step 18 decision point
The v1.54.0.0 carve moved Step 18 (documentation sync) into
ship/sections/pr-body.md and the Claude-host skeleton stopped saying
"document-release" anywhere in the workflow body — the dispatch became
invisible at exactly the moments an agent decides whether to open the
section. Restore visibility at three touchpoints, all subagent-framed
(never bare-slash-framed, which would invite an inline Skill invocation
that bypasses the fresh-context subagent + JSON contract):
- manifest trigger (renders into the section-index row AND the STOP
pointer): "dispatching the /document-release subagent to sync docs
(Step 18) and then creating or updating the PR/MR (Step 19)"
- Step 17 handoff line names Step 18's dispatch explicitly
- new hoisted doc-sync invariant beside the PR-title invariant: the
dispatch itself is never skipped; only a failed subagent is
non-blocking
Pin it in carve-guards: 'the /document-release subagent' (all three
touchpoints) + 'dispatches the /document-release subagent' (invariant)
must stay in the skeleton; the carved imperative 'Dispatch
/document-release as a subagent' must stay carved. Skeleton cap
91,600 → 92,300 (measured 91,764; trigger renders twice). Goldens
regenerated for all three hosts.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: pin the ship→document-release Step 18 wiring with a free tripwire
Five substring/structure asserts across the carved section, the Claude
skeleton's three touchpoints, the manifest trigger, and the codex/factory
goldens (inlined Step 18 ordered before Step 19). Claude-golden asserts
deliberately omitted: host-config.test.ts already enforces golden ==
generated byte-for-byte.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* test: gate-tier E2E proving /ship dispatches the document-release subagent
New skill-e2e-ship-docsync: a live agent gets the sliced Step 17→19 tail
of the generated ship skeleton in a bare-remote git fixture (Steps 0-16
"done"), under a fake HOME so the STOP pointer and the Step 18 subagent
prompt resolve to planted copies, with a stub document-release skill that
returns the empty-result JSON contract. Hard assert: an Agent/Task
tool-call matching /document-release/i exists in result.toolCalls and
precedes any `gh pr create`. Neutral prompt (no STOP-Read priming, no
document-release mention — the prompt echoes into the transcript, so
asserts read toolCalls only).
Hardening from review: throw-on-marker-drift fixture slice; per-test
GSTACK_HOME + .redact-prepush-prompted marker (routes Step 17's
credential guard to its silent branch — the hermetic GSTACK_HOME pin
defeats a HOME-only override); 480s/540s timeouts (nested subagent adds
wall clock the 300s sibling never carried); 'timeout' accepted in
exitReason only because the dispatch assert is independently hard;
whole-file describeE2ETier('gate') composed with diff selection (keeps
the file out of the periodic shard census, which sits at its ceiling,
and under the hard tier-alignment invariant).
Registered as 'ship-docsync' in E2E_TOUCHFILES + E2E_TIERS (gate) in the
same commit — touchfiles.test.ts rejects either half landing first.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: fix stale document-release TODOS entry + three review-deferred items
The SHIPPED entry still described the deleted Step 8.5 post-PR cat-delegation
design from v0.8.4; replace with the current Step 18 subagent design and its
test pins. Add the three P3 items deferred from the v1.69 plan review:
dispatch receipt enforcement, land-and-deploy→canary dispatch-pin pattern,
and the periodic shard-census boundary.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: pre-landing review fixes
Testing-specialist findings, all mechanical: (1) pin the E2E fixture's git
branch (-b main / init.defaultBranch=main) and assert every setup command's
exit status so operator git config can't silently corrupt a paid run;
(2) tighten the dispatch matcher to Step 18-prompt-specific markers
(document-release/SKILL.md | executing the /document-release workflow) so a
subagent merely quoting section text can't false-pass the regression assert
(verified against recorded burn-in transcripts); (3) replace the subsumed
carve-guards anchor with three non-overlapping per-touchpoint anchors
(gerund/imperative/3rd-person) so each touchpoint is independently enforced.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: red-team review fixes
Five informational findings: TODOS shard-census arithmetic corrected (census
is 67 with one free ungated slot; the SECOND ungated file trips the floor)
and version pointer fixed (v0.18.2.0, not v0.18.1.0); the free tripwire now
pins the two dispatch-matcher marker strings so a pr-body prompt reword
fails the free suite instead of surfacing as a paid-tier mystery; the E2E
matcher gains a section-paste exclusion (scaffold strings disqualify) —
verified against all recorded runs; the E2E header documents the tierless
test:evals invisibility tradeoff.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: adversarial review fixes
Pin the E2E matcher's two EXCLUSION markers in the free tripwire (an
unpinned 'Parent processing:' reword would silently deaden the
section-paste guard while every test stayed green); add an ordering pin
(the hoisted doc-sync invariant must sit above the pr-body STOP pointer —
presence-only anchors can't catch drift below it); plant a third
cwd-relative pr-body copy inside the fixture repo, gitignored so the agent
never tries to commit test scaffolding.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: bump version and changelog (v1.70.1.0)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* docs: CHANGELOG accuracy fixes from the doc-release review
Three factual corrections the Step 18 doc subagent caught in the fresh
v1.70.1.0 entry: 5 tripwire tests (not 6), cost floor $0.63 per the cited
eval store (not $0.59), and the visibility claim scoped to decision points
(the re-run checklist mention survived the carve). Plus the E2E header's
stale pending-burn-in note replaced with the observed numbers.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix: raise bun-polyfill subprocess budget to 60s for degraded Windows runners
The 50ms-sleep test blew the 20s budget on BOTH bun retry attempts on PR
#2700's windows-latest runner (run 32989821401) — sustained AV/runner
pressure, not just the documented cold-start. Same flake passed-on-rerun on
the prompt-token-load-reduction branch yesterday. Budget only; every
assertion still checks exact output.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(ci): run the ship-docsync gate E2E in the evals matrix + silent-skip tripwire
The evals.yml matrix is hand-enumerated and the Run step never exported
EVALS_TIER, so the new whole-file-gated ship-docsync E2E would have
self-skipped even with a row — a hollow green one layer deeper than the
documented rehomed-monolith incident. Add the e2e-ship-docsync row with a
row-level `tier: gate` property, exported as EVALS_TIER by the Run step
(empty = unset for every existing row: all readers are `=== '<tier>'` or
truthiness).
New free tripwire test/evals-workflow-matrix.test.ts ratchets the class:
matrix files must exist; gate-hosting files must have a row; whole-file-gated
matrix files must carry a matching row tier; and the burn-down lists enforce
their own cleanup. It enumerates the PRE-EXISTING holes found while wiring
this (8 gate-hosting files with no row; codex/gemini rows running zero tests;
the pty-plan-smoke row hollow since its files adopted describeE2ETier) —
tracked in TODOS as the CI gate-lane hollow-coverage burn-down.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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/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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
