d1cd04e808ff3d030fd6909fb54cd8992e1dbae2
45 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Tom Boucher
|
2f920c5af3 |
fix(#4097): evidence preservation no longer sweeps the run's own input copies into .review-diagnostics/ (#4329)
* test(#4097): failing-first regression — run input copies swept into .review-diagnostics The present_results preserve+cleanup glob `_DIAG_MD=( "$RUN_DIR"/gsd-review-*.md )` matches both lane outputs and the run's own assembled input copies (prompt, instructions, roadmap, per-plan copies, project/context/research/requirements, per-lane trimmed prompts) because both share the gsd-review- prefix. Seed the full input-copy set into the existing runWriteReviewsFlow fixture and assert (a) the diagnostics dir holds exactly the lane report + .err sidecar and no input basenames, and (b) an inputs-only run creates no diagnostics dir at all and still cleans up. Both RED against the shipped block. * fix(#4097): preserve lane output only — exclude the run's input copies from evidence The preserve+cleanup glob treated every gsd-review-*.md in RUN_DIR as lane evidence, but the workflow itself writes the run's assembled INPUTS there under the same prefix (prompt, instructions, roadmap, per-plan copies, project/ context/research/requirements, per-lane trimmed prompts). Filter _DIAG_MD by basename against that closed input set instead. Direct glob iteration + case filter — identical under bash and zsh (#4099/ #4109), nullglob-safe (#2962), and the exclusion list is closed and owned in this step: a future input basename cannot silently rejoin the evidence set. Lane reports, diagnostic stubs and non-empty .err sidecars are unchanged. Emitted-Drift-Ack-Growth: review.md — #4097 narrows the present_results evidence-preservation glob: the closed input-basename exclusion list (case filter) plus its rationale note are deliberate additions so a future input basename cannot silently rejoin the evidence set. * changeset(#4097): fixed — review diagnostics no longer sweep run input copies * changeset(#4097): backfill PR number 4329 --------- Co-authored-by: sim <sim@local> |
||
|
Dennis Alexis Valin Dittrich
|
1017898cb9 |
fix(#3771): make remediation examples non-binding and surface revision conflicts (#3916)
* fix(#3771): separate the binding property from the advisory remediation
Checker findings fused "what property failed" with "how to fix it" into a
single `fix_hint` and never marked which half binds. The checker rendered
every hint under a "must fix" heading, the orchestrators injected the issues
verbatim and ordered targeted updates, and the shared revision references
mapped each hint to a prescriptive strategy — so a contract-following planner
applied a hint literally even when a smaller mechanism satisfied the same
property, or when the hint contradicted a locked decision. There was no
channel to report that conflict, and every attempt burned a revision
iteration.
Checker side: every issue now carries a binding `required_property` (the
invariant that failed) plus its evidence and severity, and `fix_hint` is
labelled non-binding wherever it appears — including the human-facing blocker
rendering, so "must fix" unambiguously names the property and never the
example.
Planner side: revision re-checks locked decisions, capability guidance and
existing plan constraints before editing; satisfying a blocker through a
smaller valid alternative counts as addressing it; and a hint that conflicts
with any of those returns `REVISION_CONFLICT` carrying the conflict and the
alternatives considered. Orchestrators route that to user choice or the
configured plan-review convergence loop without consuming retry budget.
Also applied to the UI-spec revision loop and the gap-plan hint, and the
generic pattern's stray `suggested_fix` field name is reconciled to the
plan-checker's `fix_hint`.
Nothing legitimately binding is weakened: blockers still block, severity
still gates, iteration caps and stall escalation still fire, and required
task fields and decision coverage still hold.
Refs #3771
* test(#3771): pin the binding/advisory split across the revision chain
Locks the separation at every link that carries it: the checker's issue
schema and blocker rendering, the planner's constraint re-check and
REVISION_CONFLICT return, the generic pattern's reconciled field names, and
each orchestrator's conflict routing without retry-budget consumption. Also
pins what must not have been weakened — blockers, severity gating, iteration
caps and stall escalation.
Red against the pre-fix prose: 32 of 34 assertions fail (the 2 that pass are
the preservation checks, correctly).
Refs #3771
* chore(#3771): add changeset fragment for the remediation-binding fix
* chore(#3771): acknowledge the remediation-binding growth
Five runtime-loaded files grow: the two checkers carry the binding/advisory
split where the model reads it (a `required_property` on every dimension
example, since a schema the examples contradict teaches the examples), and
the three orchestrators carry the REVISION_CONFLICT route, which has to live
with the `iteration_count`/`revision_count` state it declines to spend.
Deletes tests/emitted-drift-acks/3172-stated-failing-direction.json: it is
fully spent on next and still owned plan-phase.md, so it walls off a key it
can no longer clear (#3078). Its removal is the documented remedy for the
duplicate-key collision, not drive-by cleanup.
* fix(#3771): close the review gaps in the conflict contract
Adversarial review (Codex, Antigravity) found four real defects in the first
pass, each confirmed against the source before acting:
- The UI checker's structured return still ordered `Fix: {exact fix required}`
and "list each BLOCK dimension with exact fix required". The dimension
examples had been marked non-binding but the rendering the researcher
actually reads had not — the same omission this issue is about.
- `ui-phase` and the canonical `revision-loop` flow incremented their counter
BEFORE dispatching the reviser, so "do NOT increment on REVISION_CONFLICT"
was unreachable prose: the iteration was already spent. The increment now
sits on the return path in both.
- The conflict gate offered "accept as-is", which is an early exit from a
still-failing blocker — a weakening the brief explicitly forbids. The three
options are now adopt an alternative / override the constraint / amend the
constraint; every one resolves the conflict. Accepting an unaddressed blocker
remains available only at the unchanged iteration-cap escalation.
- The convergence route was declarative: nothing in
plan-review-convergence.md could receive a conflict. plan-phase now records
it in REVIEWS.md — the channel that loop already consumes — convergence
refuses to declare convergence over an open entry, and routing back into a
run convergence itself started is explicitly excluded as a cycle. `quick` has
no REVIEWS.md and no phase, so its convergence branch was dead prose and is
deleted in favour of asking the user.
Also reconciles the last two drifted field names (`finding`, `affected_field`)
to the plan-checker schema, and repairs a silent no-op: the few-shot
`required_property` insertion never applied because those lines are
blockquoted, and the test's own block filter was anchored on indentation only,
so a vacuous loop passed over zero blocks. Both are fixed and the filter now
asserts it found blocks.
Refs #3771
* chore(#3771): extend the growth acknowledgment for the review round
plan-review-convergence.md joins the list: the conflict route needed a
receiving end, and it lands on the seam that loop already reads (REVIEWS.md)
rather than a new mechanism. The plan-phase, ui-phase and gsd-ui-checker
entries gain the second-pass reasoning — an executable convergence branch, the
increment moved onto the return path, and the structured return that still
ordered an exact fix.
* fix(#3771): make the conflict route bounded, ordered, and owned
Round-2 adversarial review found five more defects, each confirmed in the
source before acting:
- The convergence gate sat AFTER `gsd_run state planned-phase` and the success
banner, so a run could write and announce convergence over an unresolved
conflict. OPEN_CONFLICTS is now read from REVIEWS.md and is part of the
converged CONDITION, evaluated before any write.
- plan-phase's cycle-exclusion ("unless this run was invoked by convergence")
was not a question the orchestrator can answer at runtime. plan-phase now
never invokes convergence at all — it records the conflict when a phase
REVIEWS.md exists and resolves it with the user in-place, which removes the
cycle instead of describing it.
- Closure had no owner. plan-phase writes the row, so plan-phase strikes it
resolved; convergence only reads. An open row is a live blocker, never a
stale artifact.
- Declining to increment the counter removed the only bound on the conflict
path: an agent returning the same conflict forever would loop unattended. A
conflict naming the same `required_property` twice in a row is now a stall
and escalates through the existing gate.
- verify-work's gap-plan revision loop hands `<revision_context>` to
gsd-planner and so inherits the whole contract, but stated none of it and
could not handle the conflict return. It is now covered like the others, and
is in the test's orchestrator table.
Refs #3771
* chore(#3771): acknowledge the round-2 growth
verify-work.md joins the list — the flow the second review found missed — and
the plan-phase, ui-phase and plan-review-convergence entries gain the
round-2 reasoning: the gate moved ahead of the state write, the convergence
hand-off replaced with a runtime-checkable record-and-resolve, and the
recurrence bound that replaces the counter the conflict path stopped spending.
* fix(#3771): make the convergence gate countable and stop the conflict fall-through
Third adversarial round (Antigravity) found three defects:
- The OPEN_CONFLICTS pipeline had no `grep -v '~~'` despite its own comment
claiming one, and `grep -c '^| '` also counts a markdown table's header and
separator rows — every resolved conflict would have read as open and
convergence would have deadlocked instead of converging. plan-phase now
records each conflict as a `- [ ]` checklist line and flips it to `- [x]`, so
the gate is an exact fixed-string match with no table parsing.
- "then continue below" fell through to the checker spawn, so a SECOND
REVISION_CONFLICT would have been handed to the checker as though it were a
revised plan. plan-phase, quick and verify-work now re-evaluate the return
from the top of the conflict handler; ui-phase already looped back.
- revision-loop.md still described plan-phase routing a conflict to the
convergence loop instead of asking — the behaviour round 2 removed. Recording
is now stated as being in addition to asking, never instead of it.
Refs #3771
* chore(#3771): bring the changeset in line with what shipped
Two review rounds widened the change after the fragment was written:
verify-work's gap-plan revision and the convergence loop are covered, two
more drifted field names are reconciled, and the conflict path carries an
explicit recurrence bound.
* fix(#3771): declare and emit the REVISION_CONFLICT marker
check:contract-drift on CI caught what local lint never reached: four
workflows dispatch on `## REVISION_CONFLICT`, but no agent declared or emitted
it — an orphan consumer, matching a marker nothing produces. The shared
reference (planner-revision.md Step 7b) described the return; the agent
definitions did not carry it.
gsd-planner and gsd-ui-researcher now emit the marker in-fence alongside their
other return markers, and both registry rows in agent-contracts.md declare it.
gsd-planner's Consumed by gains the two workflows that dispatch on it and were
missing from the row.
The gate is right: a return contract belongs where the agent is defined, not
only in a reference the agent happens to load.
Refs #3771
* chore(#3771): acknowledge the return-marker growth
gsd-planner.md and gsd-ui-researcher.md each gain the REVISION_CONFLICT
marker that check:contract-drift requires them to emit.
* fix(#3771): hoist the shared conflict protocol out of the workflows
Two CI failures, both correct gates:
- tests/few-shot-calibration.test.cjs pins the plan-checker calibration file
at exactly 4 examples (2 positive, 2 negative). The example added in the
first pass broke that balance — and described PLANNER behaviour in the
CHECKER's calibration set, which is the wrong surface for it. Removed; the
smaller-alternative rule is already normative in gsd-plan-checker.md and
planner-revision.md, and pinned by the regression suite.
- tests/phase6-capstone-conformance.test.cjs (ADR-857 phase 6, #1168) requires
plan-phase.md to stay BELOW its pre-phase-6 baseline of 94519 bytes. The
inline conflict block pushed it to 94988.
The fix for the second is the one that should have been made first: the
record/resolve/close protocol and the recurrence bound were identical in four
workflows, and revision-loop.md — which plan-phase already @-imports — is what
a shared contract is for. The protocol now lives there once; plan-phase states
only its bindings (which counter, which artifact, which next step) and points
at it. plan-phase.md: 94988 -> 92739, under the ratchet with headroom, and the
four-way duplication is gone.
quick, ui-phase and verify-work do not import the reference, so they keep their
inline statements. The suite asserts each rule against what the runtime
actually loads for that orchestrator, not against the file in isolation.
Refs #3771
* docs(#3771): state the shared-protocol relationship accurately
Three of the four revision-bearing workflows do not @-import revision-loop.md,
so 'follows it verbatim' overstated the coupling. Only plan-phase defers; the
others restate the rules inline and this section is the authority they must
agree with.
* refactor(#3771): name the authority instead of restating it four times
Self-review finding: the same ~700-byte reviser paragraph was inlined in four
prompts while gsd-planner already loads planner-revision.md whenever
<revision_context> is present (agents/gsd-planner.md:555, :588) — a fifth copy
of the same contract. Four places to edit in lockstep is precisely the drift
class this PR exists to fix; the generic pattern calling the field
suggested_fix while the checker emitted fix_hint is what that looks like after
a year.
Each prompt now carries only the load-bearing clauses and names the authority
it summarises. Deliberately NOT reduced to a bare pointer: these are LLM
prompts, and a contract stated only in a file the reader is supposed to fetch
is the failure mode of this very bug. The saving is modest (~50 bytes each) —
the point is the named source of truth, not the bytes.
Suite: dropped seven assertions that pinned heading text and bold-lead
phrasing a reword would break without changing what the runtime is told; their
neighbours already pin the same contract by content. 519 -> 499 lines, 61
tests. Red gate against origin/next: 56 of 61 fail.
Refs #3771
* fix(#3771): sanitize agent-authored conflict text and bound total conflicts
Cross-AI review (agy/Gemini) raised two security findings. I confirmed the
first by running the gate's own awk against a crafted file and reproduced it
exactly; the second is a real hole in my bound.
1. FAIL-OPEN in the convergence gate. The conflict text is agent-authored and
is appended to REVIEWS.md, which the gate scans with an awk that stops at
the next '## ' heading. One line of agent text beginning '## ' ends that
scan early, so conflicts below it are never counted and convergence declares
success over a live blocker. Measured: 3 open conflicts, awk returned 2.
Fixed at the write boundary, which is the trust boundary: every field has
newlines and tabs collapsed to spaces and a leading '#', '-', '|' or fence
stripped, so one conflict is exactly one line. Both producing agents now
declare their fields single-line plain text, and the reader states the
invariant it depends on so a later edit cannot silently break it. Verified:
3 open + 1 resolved now counts 3; missing file and absent section count 0.
2. The recurrence bound was 'same required_property twice in a row', which an
agent alternating property names never trips, leaving the un-incremented
conflict path unbounded. Now bounded twice: the repeat rule catches the
common case, and the THIRD conflict return of a loop escalates whatever
property it names. A conflict still never consumes a revision iteration;
this cap is separate from and additional to the revision cap.
Rejected from the same review: deleting 'a planner that reaches
required_property by a smaller or different mechanism has addressed the issue
in full' from the CHECKER prompt as misplaced. It is load-bearing exactly
there. A checker that does not know a different mechanism counts will re-flag
the issue on re-check, which is the revision loop that never terminates. The
argument offered for deleting it, that the checker evaluates the new state
independently, describes the failure mode.
Refs #3771
* fix(#3771): fail closed on an unverifiable convergence gate
Second cross-AI pass (agy, this time with the full files rather than the diff)
found two more, both real:
1. The gate read REVIEWS_FILE with `2>/dev/null || echo 0`, so an unreadable or
empty path counted as ZERO open conflicts and converged. That path is
resolved a few lines earlier by a pre-existing unquoted
`ls ${phase_dir}/${padded_phase}-REVIEWS.md` (line 346, not touched by this
PR), which yields an empty string rather than an error when the path
contains a space. Unverifiable is not clean: the gate now tests -z and -r
first and BLOCKS. Verified both branches.
The unquoted ls itself is left alone deliberately — it predates this change
and belongs to the reviews lookup, not the conflict gate. Fixing it at my
own boundary removes its effect on this gate without widening scope.
2. REVIEWS.md is writable by the review agent, which could flip a `- [ ]` to
`- [x]` or delete the section and forge the state of a blocking gate. The
section now declares a single writer: /gsd:plan-phase appends and closes,
every other agent leaves it byte-for-byte alone, readers read.
Also trimmed a clause that explained the increment ordering by reference to
what the file said before this PR. Commit history is not instruction, and
these files are prompts.
Rejected: the claim that quick's conflict gate deadlocks autonomous pipelines
by asking the user. Its existing max-iteration escalation in the same file
already asks the user the same way; this adds no new interaction class.
Noted but out of scope: the per-dimension YAML example blocks and the shim
boilerplate duplicated across agent prompts both predate this change.
Refs #3771
* fix(#3771): count conflicts by line shape, not by section
CodeRabbit review on the rehearsal PR. Five findings, all valid, all applied.
The best one is a deletion. The convergence gate scanned between
'## Plan-Revision Conflicts' and the next '## ' heading, and that scan stops at
the FIRST heading it meets — so one stray '## ' line hid every conflict beneath
it and returned 0, converging over a live blocker. Reproduced: section-scan 0,
shape-scan 1. Sanitizing at the write boundary does not cover a hand-edited,
legacy, or foreign-written REVIEWS.md, so the reader needed its own guarantee.
It now matches the conflict line SHAPE anywhere in the file:
grep -c '^- \[ \] .*required_property:'
No section bookkeeping, nothing a heading can truncate, and it composes with the
writer's sanitization (which strips a leading '-' from agent text, so agent prose
cannot forge the shape). Verified: injected heading -> 1, all resolved -> 0.
The other four:
- Both checkers told the author never to emit a contradictory fix_hint, then
offered an escape hatch that put the forbidden route in the hint anyway. They
now name NO route in that case and state only that the property conflicts with
the constraint. A hint carrying a forbidden route is applied by anyone who
trusts hints.
- The REVISION_CONFLICT marker description in gsd-planner.md was narrower than
planner-revision.md: it covered a contradictory hint but not an unreachable
required_property. A planner reading only the agent file would have burned
retry budget on the case the reference routes to a conflict.
- The few-shot calibration examples used uppercase BLOCKER/INFO while the schema
defines blocker/warning/info. Pre-existing, but it is the same schema-vs-example
disagreement this PR exists to end, and the file was already being edited.
- verify-work's re-entry instruction existed but sat after the Bounded clause, so
the paragraph read "re-spawn ... stop re-spawning ... after re-spawning". The
re-entry now immediately follows the re-spawn, and states that only a
non-conflict return may reach the checker or increment iteration_count.
Refs #3771
* fix(#3771): resolve the contradictory scope_sanity severity examples
Sixth CodeRabbit finding, posted outside the diff range and missed on my first
read — I had claimed all findings were addressed after reading only the five
inline comments. This one was in the review body.
agents/gsd-plan-checker.md carried TWO scope_sanity examples with identical
metrics (5 tasks, 12 files) and OPPOSITE severities: warning in Dimension 5,
blocker in <examples>. Line 872 states "2-3 tasks/plan good, 4 warning, 5+
blocker" and the severity table lists warning as "Scope 4 tasks (borderline)",
so the warning example contradicted both.
ADR-2629 Decision 5's "over budget is a WARNING, never a blocker" does not
excuse it: that rule governs the smart-zone TOKEN estimate (the estimate-check
verb, lines 299-306), which is a different axis from task count. Verified in
source before touching it.
The contradiction is pre-existing but this PR made it binding and visible:
severity is now declared part of the binding payload, and both examples were
given the same required_property, so they now disagree on the severity of an
identical finding about an identical property.
Deviating from the proposed correction, which was warning -> blocker: that
would duplicate the <examples> entry outright (same tasks, files, severity).
The Dimension 5 example is instead made a genuine 4-task borderline warning, so
the file keeps one worked example per severity and the thresholds, the severity
table and both examples finally agree.
Refs #3771
* fix(#3771): stop laundering a grep error into zero open conflicts
Seventh CodeRabbit finding — from a SECOND review round my own CR-4 push
triggered, which I had not looked for. This one is a regression I introduced
while fixing the previous fail-open.
CR-4 replaced the truncatable section scan with:
OPEN_CONFLICTS=$(grep -c '^- \[ \] .*required_property:' "$REVIEWS_FILE" || true)
`|| true` masks every grep failure. grep exits 1 for "no matches" (a legitimate
zero) but 2 for a read error, and `|| true` turns both into an empty capture
that `${OPEN_CONFLICTS:-0}` renders as 0. If REVIEWS.md is removed or becomes
unreadable between the -r check and the scan, the gate reports no conflicts and
convergence proceeds. Proven: unreadable file -> captured empty -> 0.
The status is now inspected, and only exit 1 counts as zero; anything else
blocks.
My first attempt at this fix was itself wrong and my own harness caught it: I
wrote `if ! grep ...; then grep_status=$?`, but `!` inverts the status, so `$?`
in that branch is 0 and every failure reads as success — the clean-file case
printed "BLOCKED (grep exit 0)". The status must be read in the ELSE branch of a
non-negated `if`, which is what CodeRabbit proposed. Both traps are now pinned
by tests.
Verified end to end: all resolved -> 0, no conflicts at all -> 0, injected
heading -> 1, unreadable file -> BLOCKED with grep exit 2.
Refs #3771
* test(#3771): execute the conflict gate instead of reading it
CodeRabbit round three: 0 actionable, 1 nitpick — "these assertions inspect
Markdown source only; they do not prove that grep status 1 produces zero
conflicts or that a scan error exits before convergence." Rated Trivial. It is
the most valuable finding of the three rounds.
This gate has been wrong three times: a section scan a heading could truncate, a
`|| true` that laundered grep's error status into zero, and an `if !` whose `$?`
reported the negation rather than the command. Every one of those passed the
text assertions that existed at the time. I proved each fix by hand in a shell,
and none of that proof lived in the suite.
The gate is one self-contained fenced block, so the test now extracts it from
the workflow — located by content, not line number — writes it to a script and
RUNS it against fixtures: two open plus one resolved counts 2; no matches counts
0 and does not fail; a conflict below an injected `## ` heading still counts; an
unreadable path and an empty path both BLOCK with a non-zero status and no zero
count on stdout.
Non-vacuity proven by mutation rather than asserted. Reverting the gate to each
of its three historical broken forms reds the suite:
section-scan awk -> 7 failures (5 in the gate cases)
|| true -> 4 failures (3 in the gate cases)
if ! (negated $?) -> 4 failures (3 in the gate cases)
restored -> 69 pass, 0 fail
The prose assertions stay: they are the right instrument for a prompt. This
covers the one part of the change that is real shell an orchestrator executes.
Refs #3771
* test(#3771): route the gate harness through the shared test helpers
ESLint's project rules caught three violations in the new harness: an unbounded
execFileSync (DEFECT.UNBOUNDED-SUBPROCESS — an unbounded spawn is an indefinite
hang, and on macOS CI that is how a stuck run stops reporting instead of failing)
and two raw fs.rmSync calls, which skip the Windows-EBUSY retry budget that
helpers.cleanup carries.
Now uses createTempDir/cleanup from tests/helpers.cjs and passes an explicit
30s timeout. Suppressing the rules was available and would have been the wrong
call: both exist because of real CI failure modes on platforms I am not testing
on.
* chore(#3771): backfill the changeset PR number
The pr: field is drift-checked against the PR event payload, so it cannot be
written before the PR exists. Set to 3916.
* fix(#3771): close revision conflict persistence gaps
Use the authoritative review artifact, keep conflict and normal retry paths disjoint, and enforce one writer-reader grammar so malformed state fails closed.
Emitted-Drift-Ack-Growth: diagnose-issues.md — #3771 marks the gap-plan remediation hint non-binding while keeping root_cause authoritative
Emitted-Drift-Ack-Growth: gsd-plan-checker.md — #3771 separates binding required_property evidence from advisory fix_hint examples across the checker contract
Emitted-Drift-Ack-Growth: gsd-planner.md — #3771 declares the REVISION_CONFLICT return used when remediation contradicts governing constraints
Emitted-Drift-Ack-Growth: gsd-ui-checker.md — #3771 applies the same binding-property and advisory-hint split to UI review findings
Emitted-Drift-Ack-Growth: gsd-ui-researcher.md — #3771 defines the UI revision producer's structured REVISION_CONFLICT return
Emitted-Drift-Ack-Growth: plan-phase.md — #3771 routes and persists bounded revision conflicts before spending the normal retry budget
Emitted-Drift-Ack-Growth: plan-review-convergence.md — #3771 adds the fail-closed owned-block parser and prevents convergence over open conflicts
Emitted-Drift-Ack-Growth: review.md — #3771 emits and preserves the canonical writer-owned conflict block across review regeneration
Emitted-Drift-Ack-Growth: ui-phase.md — #3771 routes UI revision conflicts to resolution before consuming revision_count
Emitted-Drift-Ack-Growth: verify-work.md — #3771 gives gap-plan revision the same bounded conflict route before iteration_count
* test(#3916): guard rebases against schema drift
Load the current-base progressive-disclosure examples so every integrated issue remains bound by required_property after branch reconciliation.
* test(#3916): skip the extracted-gate suite's bash spawns on win32
Third review round's sole survivor: runConflictGate()/withReviews() spawn
bash against a Node-native temp path built by createTempDir(), which is
backslash-separated on the Windows CI lane and not a path Git Bash is
guaranteed to accept (DEFECT.WINDOWS-TEST-PORTABILITY, matching the
observed CI failure at revision-remediation-binding.test.cjs:844,
ENOENT on a path Windows read as a directory separator). No eslint rule
catches it since the call has neither a chmod nor a `bash -c` form.
Guards the four call sites with the repo's existing skipOnWin32
convention (describe/test `{ skip: IS_WINDOWS }`) rather than
normalizing the harness path to forward slashes, which would defeat the
one test whose purpose is proving the production gate does NOT rewrite
a literal backslash in a POSIX filename.
* fix(#3916): backfill changeset pr field to the fork validation PR number
* fix(#3771): forbid silently accepting an open plan-revision conflict at max-cycles escalation
The max-cycles escalation prompt only surfaced HIGH_COUNT and ACTIONABLE_COUNT; an open
plan-revision conflict (OPEN_CONFLICTS > 0) was never disclosed there, and "Proceed anyway"
could exit successfully over it — exactly the failure mode this PR exists to close (a success
banner over an unresolved conflict nobody resolved). Blockers still block: withhold "Proceed
anyway" and route to Manual review whenever a conflict is open.
* fix(#3771): do not hard-block REVISION_CONFLICT persistence when no REVIEWS.md exists yet
A phase's first-ever revision cycle can return REVISION_CONFLICT before any REVIEWS.md has been
written — REVIEWS_PATH is then legitimately empty, not a corrupt or deleted file. The persistence
gate's own accompanying prose already says the record channel applies 'when REVIEWS_FILE is
non-empty', but the bash condition never checked that, so it hard-blocked every conflict on a
brand-new phase regardless of whether persistence was even expected to run. Require a non-empty
REVIEWS_FILE before treating a missing file as an error.
* chore(#3771): raise the plan-phase.md ADR-857 host-loop ceiling to 96700
The frozen pre-phase-6 ceiling (94519) collided on rebase: this PR's own
REVISION_CONFLICT persistence/routing gate is core planner control flow, not an
un-extracted optional feature, and landed alongside an unrelated, already-merged
same-file growth (the #4.6 context-drift pre-check) already on next. Same
rationale #1298 already established for execute-phase.md's ceiling.
* chore(#3916): backfill changeset pr field to the upstream PR number
* fix(#3771): make the writer-side REVISION_CONFLICT sanitize step real shell
The Conflict Return record channel sanitized agent-authored fields via a
prose instruction ("Sanitize each agent-authored field before appending")
for the orchestrator LLM to apply by hand, while the reader-side gate in
plan-review-convergence.md parses the same slot with real, executed awk.
Flagged Minor across two review rounds (round 4, round 6) since no code
performed the sanitize anywhere.
plan-phase.md's Conflict Return step now runs a real bash gate: sanitize
each field (collapse newline/tab to space, strip a leading #/-/|/fence),
build the one-line record, skip the append if an identical line already
exists (idempotent), insert before the writer-owned end delimiter, and
fail closed if that delimiter is missing rather than silently dropping
the conflict.
tests/revision-remediation-binding.test.cjs extracts and RUNS the new
fence (matching how the reader gate is already tested), composing it
with the existing reader gate: hostile-field fast-check fuzzing, a
repeated-conflict idempotency check, and a missing-delimiter fail-closed
check that the file is left byte-for-byte unchanged on failure.
Emitted-Drift-Ack-Growth: plan-phase.md — #3916 turns the writer-side
REVISION_CONFLICT sanitize+insert step into real, executed shell instead
of a prose instruction, matching the reader gate's existing rigor
* fix(#3916): backfill changeset pr field to the fork validation PR number
Fork CI's changeset-lint reads the real PR number from its own event
payload; the fragment still carried the upstream number from the last
sync, so the DEFECT.CHANGESET-PR-FIELD-DRIFT check failed on this fork
PR. Re-backfill to the upstream number before the final push.
* fix(#3771): close the awk -v forgery and same-session close gaps agy found
Adversarial review (gemini-3.8-flash-high via the internal agy review
lane) on the full PR found two BLOCKERs against the just-added
writer-side conflict gate:
1. `awk -v line="${LINE}"` decodes escape sequences in its argument, so
a literal two-character `\n` in agent-authored text became a real
newline inside awk, splitting the appended record across two
physical lines. `tr` only strips actual control bytes, so it never
saw this — it defeated the exact forgery the gate exists to
prevent, both the reader's zero-count and the writer's own
idempotency check. Fixed by passing LINE/END through awk's
ENVIRON, which is not escape-decoded.
2. A conflict resolved and re-spawned within the same plan-phase
session was never flipped from `- [ ]` to `- [x]` — the record
channel bullet said "plan-phase closes it," but no step did. Only
a *separate* `--reviews` re-entry (line ~622, still prose-only)
closes conflicts; the in-session resolve path left them open
forever, permanently blocking convergence. Fixed by carrying the
just-written line in `PENDING_CONFLICT` and closing it in the
`Otherwise` branch before the checker re-spawns.
Also fixed a MAJOR: docs/COMMANDS.md described the `--max-cycles`
escalation gate as uniformly offering "proceed or review manually,"
but the code (this PR's own change) withholds "Proceed anyway"
specifically when a plan-revision conflict is open — only manual
review is offered in that case. Docs now say so.
Not applied: the reviewer's `\r` truncated to plain tr from a MINOR
that also asked for temp-file permission preservation across `mktemp`.
Applying `chmod --reference` is not portable to macOS/BSD `chmod`, so
this is left as a documented low-severity tradeoff — the temp file now
sits alongside REVIEWS.md (same filesystem, atomic `mv`), which was
the same finding's more substantive half. Also not applied: a
suggested `gsd_run review record-conflict` CLI subcommand to
deduplicate the two `awk` blocks — a new command plus wiring is out of
scope for a review-remediation fix.
tests/revision-remediation-binding.test.cjs adds regression coverage
for both BLOCKERs: a literal-backslash-n hostile field composed with
the reader gate, and a close-gate extraction that verifies the flip to
`[x]`, the reader's count dropping to 0, and a fail-closed path when
the pending line is missing.
Emitted-Drift-Ack-Growth: plan-phase.md — #3916 fixes an awk -v escape-
decoding forgery and adds the missing same-session conflict-close step
an adversarial review found in the writer-side gate
* chore(#3916): backfill changeset pr field to the upstream PR number
Fork-validation CI needed pr: 1 to pass its own changeset-lint; restore
pr: 3916 before this push reaches open-gsd/gsd-core.
* fix(#3771): trim plan-phase.md prose back under the XL byte cap
Merging origin/next's unrelated growth pushed plan-phase.md 473 bytes
past the workflow-size-budget XL cap and the ADR-857 phase-6 baseline,
both tripped by CI after review approval. Removed an unpinned inert
bash comment and tightened connective prose in three REVISION_CONFLICT
bullets; no executable shell or test-pinned substring changed.
* chore(rehearsal): pin changeset pr field to fork rehearsal PR #25
Scratch-only commit for the rehearsal branch's own CI. Will not be
carried onto the branch backing upstream #3916 — that keeps pr: 3916.
* fix(#3771): address CodeRabbit findings on the REVISION_CONFLICT protocol
Fork rehearsal PR #25's first CodeRabbit pass surfaced 7 findings against
the already-approved #3916 diff; each verified against current code
before fixing (none hallucinated):
- plan-phase.md: writer-side awk gates now strip a trailing \r before
comparing lines, matching the reader gate (plan-review-convergence.md)
-- a CRLF REVIEWS.md previously made both writer gates fail closed.
- plan-phase.md: the close-fence's REVIEWS_FILE/PENDING_CONFLICT/
CONFLICT_RESOLUTION were read without ever being (re)defined in that
fence -- shell state does not survive across separate fenced blocks
(same convention already documented in review.md). Added the explicit
recompute/set instruction.
- revision-loop.md: previous_conflict_property was never reset after a
normal (non-conflict) revision, so a later, unrelated conflict on the
same property could be misread as a repeat and escalate prematurely.
- gsd-plan-checker.md / few-shot-examples/plan-checker.md: two example
required_property strings were unconditionally binding in a way their
own dimension's rules aren't (no-analog RESEARCH.md fallback; tasks
that create no functions), now scoped to match.
- quick/steps/plan-checker-loop.md: added the same disjoint
"Otherwise (not REVISION_CONFLICT)" branch plan-phase.md already had,
closing an ambiguity between the conflict and non-conflict return paths.
- revision-remediation-binding.test.cjs: the REVIEWS_PATH init-order
assertion used indexOf() without checking for -1, so it would pass
vacuously if either anchor were renamed away.
Also restores an "Export the row's CONFLICT_*" instruction I had cut in
the prior byte-budget trim -- checked non-pinned by tests, but it was the
only text telling the agent to set those vars before the awk block reads
them via ENVIRON.
Net growth from these fixes required reclaiming bytes elsewhere in
plan-phase.md (verified against every pinned substring in
revision-remediation-binding.test.cjs) to stay under the XL tier's
hard 98304-byte cap; final size 98245 bytes.
* fix(#3771): resync the #4079 shrink-only mirror to the current PRE_PHASE6 line
tests/plan-phase-background-wait-wakeup.test.cjs (landed on next via an
unrelated #4079 PR, merged in by this branch's next-sync) mirrored
plan-phase.md's phase6 shrink-only ceiling as a hardcoded local constant
(94519) rather than reading tests/phase6-capstone-conformance.test.cjs's
PRE_PHASE6 value. That value has since been legitimately raised twice
during this PR's own review (94519 -> 96700 -> 98300) to accommodate the
REVISION_CONFLICT persistence/routing gate. The two branches' independent
histories left the mirror stale post-merge -- not a textual git conflict,
but the same class of thing. Resynced to 98300.
* fix(#3771): address round-2 CodeRabbit findings on the conflict gates
CodeRabbit's re-review of the previous remediation commit found two real
issues in what it had already flagged:
- Both writer-side awk CRLF fixes used \`sub(/\r$/, "")\` directly on \`\$0\`,
which mutates it in place -- \`{ print }\` then emitted the CR-stripped
copy for every passed-through line, silently rewriting an unrelated
CRLF REVIEWS.md to LF on any insert or close. Now compares against a
separate \`cur\` copy and prints the original, untouched \`\$0\`.
- The close-fence's "recompute REVIEWS_FILE/PENDING_CONFLICT" prose
implied in-fence derivation, but the fence has no such code and the
test harness (\`runCloseGate\`) deliberately supplies all three as
pre-set env vars -- matching how the open fence's "Export the row's
CONFLICT_*" instruction already works. Reworded to "export ... in the
same invocation", matching that established, test-verified pattern
instead of promising logic that isn't there.
Added a regression test proving the CRLF fix no longer touches
passthrough lines (red against the mutate-in-place version, green now).
* fix(#3771): use a CRLF-safe check in the new passthrough regression test
local/no-crlf-fragile-split forbids splitting readFileSync content on a
literal \n (Windows git-autocrlf checkouts yield \r\n). My CRLF
passthrough-preservation test from the previous commit did exactly that
to inspect the first line. Replaced with a direct startsWith() check
against the known CRLF-terminated header, which needs no split.
* test(#3771): assert the record itself is inserted in the CRLF passthrough test
CodeRabbit nitpick (round 3): the passthrough-preservation test checked
gate status and the pre-existing line's CRLF ending, but never asserted
the new REVISION_CONFLICT record was actually written.
* fix(#3771): address agy/gemini-3.8-flash-high adversarial review findings
Full-PR adversarial review (internal /gsd-review antigravity lane,
gemini-3.8-flash-high) surfaced 9 findings; each verified against current
code before fixing (none hallucinated):
HIGH:
- quick-batch/steps/plan-checker-loop.md never received the
required_property/fix_hint binding language or REVISION_CONFLICT
handling this PR added everywhere else -- a genuinely unmigrated
producing context. Migrated to match quick/steps/plan-checker-loop.md,
and added it to the ORCHESTRATORS consistency battery in
revision-remediation-binding.test.cjs so future drift is caught
automatically.
- The close-fence's PENDING_CONFLICT was an agent-supplied env var that
had to exactly reconstruct a five-field sanitized line across a
multi-minute subagent dispatch -- fragile, and a scalar var also meant
a second simultaneous conflict silently dropped the first on overwrite.
Redesigned to match the open conflict by CONFLICT_DIMENSION/
CONFLICT_PLAN identity instead: the agent re-supplies two short,
already-tracked identifiers rather than reconstructing the full
sanitized text, and each conflict resolves independently regardless of
how many are open. Updated the test harness's runCloseGate contract to
match, and added a two-open-conflicts regression test.
MEDIUM:
- plan-phase.md's `--reviews` replanning path told the reader to "flip
the matching line to [x]" in prose only, with no executable path to
it -- pointed it at the same close gate used in step 12.
- plan-review-convergence.md's reader-gate awk tolerated a blank line
before the opening delimiter but not before the heading that follows
it; a formatter or LLM writer inserting one would hard-abort
convergence on an otherwise well-formed REVIEWS.md. Added the same
tolerance already granted above it, with a regression test.
LOW:
- Clarified that the escalation destination for a stalled conflict is
the same iteration/revision-count cap gate already defined in each of
quick, quick-batch, ui-phase, and verify-work, rather than an
undefined "stall" concept.
- Clarified "twice in a row" means no successful revision intervened,
matching revision-loop.md's now-explicit previous_conflict_property
reset.
- Fixed gsd-ui-researcher.md's stale rationale text, copied verbatim
from planner-revision.md: ui-phase presents the conflict table
directly to the user, it does not persist to a shared file scanned by
heading.
Net growth again required reclaiming bytes in plan-phase.md (verified
against every pinned substring in revision-remediation-binding.test.cjs)
to stay under the XL tier's hard 98304-byte cap; removed a now-dead
PENDING_CONFLICT assignment in the process. Final size 98258 bytes.
* fix(#3771): scope row 48's quick/steps guard away from plan-checker-loop.md
tests/gsd-quick-batch-quick-regression.test.cjs's row 48 (#3676) flagged
this branch's quick-batch/steps/plan-checker-loop.md migration (the agy
HIGH finding) as a violation, because it also edits
quick/steps/plan-checker-loop.md for the same underlying #3771 protocol
fix.
Verified against git history before scoping:
|
||
|
Behruz Nassre Esfahani
|
5ad9a36f35 |
fix(#4255): resolve reviewer-lane effort from the lane, not from gsd-plan-checker (#4275)
`review-lane plan` resolved every cross-AI reviewer lane's reasoning effort by spawning `query resolve-execution gsd-plan-checker --host <slug>`. The agent id was a hardcoded literal, so `--host` chose only the argv RENDERING while the LEVEL always came from the installed plan-checker's frontmatter — `low` under every shipped model profile. Every prompt-fed lane therefore ran at a fast structural verifier's effort, and because the rendered argument is a CLI config override it silently beat the effort the operator had configured for that CLI. At `low` a large source-grounded prompt makes a model end its turn with no final message, so the lane came back empty and its stub read as a crash. Effort is a property of the review, so the lane declares it. Two new fields on ReviewerLane — `effortConfigKey` (`review.effort.<slug>`) and `defaultEffort` — carried through each capability manifest and the generated registry, set on the three lanes with an argv effort channel and null on the other nine. A new pure `resolveLaneEffort()` resolves config key -> lane default -> nothing, where "nothing" emits no effort argument at all and the reviewer CLI's own configuration decides; `inherit` selects that path explicitly and an unrecognized level falls back to the lane default rather than being forwarded to a CLI that would reject it. The host's negotiated effortSurface still gates the rendering, so ADR-1239/#2481's trust boundary holds on this path too. Resolving in-process also removes up to twelve subprocess spawns per review. The empty-output stub now names the effort the lane ran at and distinguishes a clean exit from a timeout kill, a non-zero exit, and a process that never ran — `status` is null for both a timeout and a signal, so those were indistinguishable before. The hint is hedged: a clean empty exit is most often a model stopping short, but it is also consistent with a CLI writing its output elsewhere. Also: the capability validator now knows both fields, rejects a malformed key or an out-of-vocabulary default, and rejects a default declared without a config key (a level the operator could never override). An existing end-to-end row in tests/effort-surface-axis.test.cjs asserted the old coupling; it now configures the lane's own key and pins the decoupling in the same real spawn, with the agent execution tier set to a level that must not appear. Emitted-Drift-Ack-Growth: review.md — the effort/model resolution-order table this fix adds. The workflow is where an operator looks to find out which knob set a lane's model and effort; leaving the new key undocumented there is the same invisibility that made the plan-checker coupling survive this long. Emitted-Drift-Ack-Growth: review.md — the effort/model resolution-order table this fix adds. The workflow is where an operator looks to find out which knob set a lane's model and effort, so leaving the new key undocumented there is the same invisibility that let the plan-checker coupling survive. Claude-Session: https://claude.ai/code/session_01CRMEuzNMWn3gs5uUW2ghcF Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-authored-by: Tom Boucher <trekkie@nomorestars.com> |
||
|
JusticeWay
|
0fca71eaae |
enhance(#2529): cover every workflow with response-language directives + CI lint (#2558)
* enhance(#2529): cover every workflow with response-language directives + CI lint Every workflow now carries response-language coverage in one of three forms, and a CI lint keeps it that way. - 43 workflows load the new shared reference, `gsd-core/references/response-language-directive.md`, by eager `@`-import. - Lazy-loaded modes/steps/templates, which cannot rely on an eager import, carry an exact inline directive; 35 such paths are pinned by exact path. - Fragments dispatched by a covered parent inherit coverage, proven per file rather than granted per directory. The 45 workflows whose directive covered only "questions, prompts, and explanations" now name inter-tool narration, which is the defect #2529 reports: the running commentary between tool calls stayed English while the answers around it were translated. `scripts/lint-response-language-coverage.cjs` enforces it and fails closed on three independent discovery failures (unreadable catalog, empty catalog, unfollowed symlink). It resolves which reference a workflow imports and applies the same four-predicate test to that file, so a weakened shared reference uncovers its importers instead of passing silently, reported once as a systemic failure rather than 43 times. The walk follows symlinked subtrees with a realpath cycle bound. `lint:ci` invokes it by name. REQ-LANG-03 and REQ-LANG-04 state the contract in docs/FEATURES.md; REQ-LANG-04 names the two forms that satisfy it ("narration", "between tool calls") rather than enumerating class members an author cannot use verbatim, and a test pins that text to what the matcher accepts. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2529): register the coverage test in the docs-guard lane `107eb8c1` (#3787) landed the docs-guard lane on `next` while this PR was open: a test that reads a `docs/` path must be named in `scripts/docs-guard-registry.cjs` or carry a `docs-guard-exempt` marker, so the guards that read a doc run on the PR that changes it. `tests/response-language-coverage.test.cjs` reads `docs/FEATURES.md` -- it extracts every form REQ-LANG-04 offers an author and runs each through the matcher that enforces it. Registration, not exemption, is the correct side of that gate: a reword of the requirement with no code change is precisely the diff this test exists to catch, and it is the diff the lane would otherwise skip. Registered narrowly (`['docs/FEATURES.md']`) rather than with the `'*'` sentinel, so an unrelated docs change does not pull this test into the lane. Verified: lint-docs-guard-registration 0 violations, tests/ci-docs-guard-registry.test.cjs 51/51, lint:ci exit 0. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2529): consolidate this PR's emitted-growth acks into its own fragment This PR ripples emitted bytes across 85 workflow paths. Until now each ripple was acknowledged by appending to whichever live fragment owned that path, because two ack sources may never name the same path. `a84f7563` (#3078) swept all 45 fully-spent fragments off `next`. Forty-two of the paths this PR grows were owned by swept fragments, so those keys are now unowned and this PR's own fragment declares them directly -- one path, one source, and no dependence on a fragment that no longer exists. Each adopted entry keeps its measurement and records where it came from. Two paths are handled differently, because the sweep did not free them: - `review.md` is now owned by `3034-parallel-reviewer-lanes.json`, which landed on `next` after the sweep. Its entry is live, so the old route still applies: this PR's note is appended to that entry rather than declared a second time. - `plan-review-convergence.md` keeps the arrangement made in round 24. Result: 3 fragments in the directory, 85 keys in this PR's own, 0 cross-source duplicates. `lint-emitted-drift-ack` exit 0, `tests/emitted-attribution.test.cjs` green. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2529): move REQ-LANG-03/04 into the feature fragment that now generates them `36375513` (#3845) made docs/FEATURES.md a generated projection of docs/features/*.md, marked "do not edit by hand". This PR wrote REQ-LANG-03 and REQ-LANG-04 straight into the generated file, so the rebase left the requirement present in the projection and absent from its source -- the next regeneration would have deleted both, and `tests/features-index-gate.test.cjs` was already red on the mismatch. Both requirements now live in docs/features/response-language-config.md alongside REQ-LANG-01 and -02. Regenerating produces a docs/FEATURES.md that is byte-identical to the committed one, so the text this PR shipped is unchanged -- only its source of truth moved to where #3840 put it. The docs-guard registration is widened to name the fragment as well as the projection. The requirement's source is the fragment now, and an edit there that skips regeneration would otherwise reach this guard through neither path. Verified: features-index-gate 68/68, lint-docs-guard-registration 0 violations, ci-docs-guard-registry + response-language-coverage 142/142. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2529): hand the plan-phase ack back to its new live owner `c933184b` (#3825) landed `3172-stated-failing-direction.json` on `next` after fragment had adopted that path when the sweep left it unowned, so the merged tree named it from two sources -- a hard failure in `scripts/lint-emitted-drift-ack.cjs`. The path has a live owner again, so the append route applies: this PR's note joins that entry, carrying its own measurement, and the key is dropped from this PR's fragment (84 keys left, the others untouched). The provenance sentence written for the swept-fragment case is removed rather than reused -- this path was never orphaned, so that account of it would be false. Same shape as `review.md` and `plan-review-convergence.md`: ownership is a property of the merged tree, and a fragment landing upstream after a push can reclaim a key no local check would have flagged. Verified: lint-emitted-drift-ack exit 0, lint:ci exit 0. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2529): state byte figures that are true against the tree The reference claimed `execute-phase.md` has "2 bytes of headroom under the ceiling named below". That was true when the sentence was written -- the file sat at 93398 against the 93400 comfort assert -- and upstream has since shrunk it to 91493 against a 93600 hard ceiling, so the figure now understates the headroom by three orders of magnitude. The rationale the sentence supports does not depend on the number, so the number is gone rather than refreshed: a restated figure would go stale again on the next upstream edit, and nothing parses it. Audited every other numeric claim this PR ships the same way, mechanically against the merge base: all 82 FILE-delta claims in the ack fragment match the real per-file delta exactly, and the 1,629-byte reference and 63-byte import line check out. One class was imprecise: the 41 notes for workflows whose inline directive was rewritten in place quoted the conversion counterfactual as "+1,692 bytes more loaded context", which is the reference form's whole weight, not the increase over the inline directive those files already carry. Each now names both quantities and the net (+1,605 / +1,609 / +1,584). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2529): one rule for pinned vs inherited coverage, and the docs to pick it Review measured that 14 of the 35 pinned fragments would pass by inheritance anyway, and that the PR asserted both readings at once: inheritance is real coverage (so those 14 pins are noise) or it is not (so 30 inheriting fragments are green-but-uncovered). Only one can be true. Inheritance is real: the predicate proves it per file -- the parent must dispatch this exact path from a read/execute context AND be covered itself -- so the parent's directive is in the loaded context by the time the fragment is read. The 14 pins are therefore removed along with the directive lines they pinned, and those files inherit like the 30 structurally identical ones. The rule is now stated where the set is declared, and enforced from the other side by a test: no member of the pinned set may be one that would have inherited. That is what decides the form for the next fragment. - pinned set 35 -> 21; 14 workflow files revert to their base content - `findViolations` no longer returns early on a pinned path: a file that becomes eagerly loaded and takes the shared reference is strictly better off, and the gate must not red that. The reference form is admitted because its own wording is validated in turn; an arbitrary reworded inline line still fails. - the reference-directive cache is keyed by size and mtime, not by path alone, so a rewritten reference re-asked in one process no longer returns the stale verdict - `carriesInlineDirective` names its negation blindness: four independent hits read vocabulary, not polarity - the real-tree scan asserts each source produced files instead of `> 152`, a constant that read as the workflow count and would have passed a scan that lost one of its two directories - the pinned-set size assertion goes the same way: the size follows from the rule, so the rule is what the suite asserts Docs, for the gate that now governs every future workflow: - `docs/contributing/response-language-coverage.md` -- why the narration class is the discriminator, the four coverage forms, the decision order that picks one, the pinned line, and what each failure message means - a row in CONTRIBUTING.md's CI checks table, matching the docs-guard row - `docs/CONFIGURATION.md` points at it from the `response_language` entry Also: the changeset said 45 reworded workflows; it is 44 (42 @-reference + 21 pinned + 44 rewritten = 107 touched). That text ships to CHANGELOG.md. `3707-parse-gap-reporting.json` landed on `next` reclaiming `audit-uat.md` and `progress.md`; both handed back by the append route, leaving 82 keys here. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2529): correct the reference-taker count, 43 -> 42 The ack notes said the import line is byte-identical "in each of the 43 workflows that take the reference" and that the alternative would be "43 inline copies". The shared reference has 42 importers; the 43rd file in review's table is `execute-phase.md`, which imports the OTHER reference. Corrected in all 41 notes that carry the sentence, across this PR's fragment and the two it appends to. Found by re-running the numeric audit from the previous round after the rebase, which also re-verified all 84 FILE-delta claims against the new base -- all exact. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2529): migrate the emitted-drift ack from a fragment to commit trailers ADR-3942 (#3954) landed while this PR was open: the acknowledgment is now a commit trailer and tests/emitted-drift-acks/ no longer exists. The fragment is deleted and each key it declared becomes one trailer, reasons unchanged. The four keys this PR had handed to 3034-*, 3172-* and 3707-* under the one-source rule come home here. That rule was the whole reason for the hand-backs, and the trailer model has no shared namespace to collide in -- five of this PR's rounds were spent on exactly those collisions. Emitted-Drift-Ack-Growth: add-backlog.md — #2529 — RESTATED in round 10, superseding this PR's earlier "+63 bytes, prose only" wording, which reported a file delta as if it were the whole cost. The workflow gains the shared response-language directive as a single `@`-reference line. FILE delta: +63 bytes, byte-identical in each of the 42 workflows that take the reference. LOADED-CONTEXT delta: +1,692 bytes per workflow — the 63-byte import line plus the 1,629 bytes of `gsd-core/references/response-language-directive.md`, because an `@`-import in this repo is EAGER (ADR-1610 Decision point 4; docs/ARCHITECTURE.md: moving prose into a file that is still eagerly `@`-imported "shrinks the measured file without shrinking loaded context"). The repo's size gates — the tier hard caps in `tests/workflow-size-budget.test.cjs` and this size ratchet — measure the FILE, not the transitive inline, so they see 63 of those 1,692 bytes; the remaining 1,629 are declared here because no gate reads them. The eager import is accepted on its merits, not hidden: 42 inline copies would be 43 places for the wording to drift, and the reference is the one place it is maintained. Prose only: no step, gate, tool invocation, or subagent dispatch shape changed. Emitted-Drift-Ack-Growth: add-phase.md — #2529 — RESTATED in round 10, superseding this PR's earlier "+63 bytes, prose only" wording, which reported a file delta as if it were the whole cost. The workflow gains the shared response-language directive as a single `@`-reference line. FILE delta: +63 bytes, byte-identical in each of the 42 workflows that take the reference. LOADED-CONTEXT delta: +1,692 bytes per workflow — the 63-byte import line plus the 1,629 bytes of `gsd-core/references/response-language-directive.md`, because an `@`-import in this repo is EAGER (ADR-1610 Decision point 4; docs/ARCHITECTURE.md: moving prose into a file that is still eagerly `@`-imported "shrinks the measured file without shrinking loaded context"). The repo's size gates — the tier hard caps in `tests/workflow-size-budget.test.cjs` and this size ratchet — measure the FILE, not the transitive inline, so they see 63 of those 1,692 bytes; the remaining 1,629 are declared here because no gate reads them. The eager import is accepted on its merits, not hidden: 42 inline copies would be 43 places for the wording to drift, and the reference is the one place it is maintained. Prose only: no step, gate, tool invocation, or subagent dispatch shape changed. Emitted-Drift-Ack-Growth: add-tests.md — #2529 MAJOR 1 (round 10): the workflow's pre-existing inline response-language directive is rewritten IN PLACE so the sentence names inter-tool NARRATION explicitly — narration between tool calls, status updates, progress notes, findings — instead of only "questions, prompts, and explanations". That older wording is the defect #2529 reports (it leaves the running commentary between tool calls in English while the answers around it are translated), and `scripts/lint-response-language-coverage.cjs` had been certifying it as coverage, so the gate legitimised the bug. +87 bytes, prose only: no step, gate, tool invocation, or subagent dispatch shape changed. FILE delta and LOADED-CONTEXT delta are both +87 here, and that identity is the point — the directive was deliberately NOT converted to an `@`-reference, because an `@`-import in this repo is EAGER (ADR-1610 Decision point 4; docs/ARCHITECTURE.md: moving prose into a file that is still eagerly `@`-imported "shrinks the measured file without shrinking loaded context"), so the conversion would have bought a smaller measured file at a cost of 1,692 bytes of loaded context per workflow — the 63-byte import line plus the 1,629-byte reference — against the 87 bytes this inline directive costs, a net +1,605. Stated plainly because the gates cannot state it: the repo's size gates — the tier hard caps in `tests/workflow-size-budget.test.cjs` and this size ratchet — measure the FILE, not the transitive inline, so an `@`-reference conversion would have READ as a smaller change to every gate in the repo while costing 1,605 bytes more loaded context per invocation. Emitted-Drift-Ack-Growth: add-todo.md — #2529 MAJOR 1 (round 10): the workflow's pre-existing inline response-language directive is rewritten IN PLACE so the sentence names inter-tool NARRATION explicitly — narration between tool calls, status updates, progress notes, findings — instead of only "questions, prompts, and explanations". That older wording is the defect #2529 reports (it leaves the running commentary between tool calls in English while the answers around it are translated), and `scripts/lint-response-language-coverage.cjs` had been certifying it as coverage, so the gate legitimised the bug. +87 bytes, prose only: no step, gate, tool invocation, or subagent dispatch shape changed. FILE delta and LOADED-CONTEXT delta are both +87 here, and that identity is the point — the directive was deliberately NOT converted to an `@`-reference, because an `@`-import in this repo is EAGER (ADR-1610 Decision point 4; docs/ARCHITECTURE.md: moving prose into a file that is still eagerly `@`-imported "shrinks the measured file without shrinking loaded context"), so the conversion would have bought a smaller measured file at a cost of 1,692 bytes of loaded context per workflow — the 63-byte import line plus the 1,629-byte reference — against the 87 bytes this inline directive costs, a net +1,605. Stated plainly because the gates cannot state it: the repo's size gates — the tier hard caps in `tests/workflow-size-budget.test.cjs` and this size ratchet — measure the FILE, not the transitive inline, so an `@`-reference conversion would have READ as a smaller change to every gate in the repo while costing 1,605 bytes more loaded context per invocation. Emitted-Drift-Ack-Growth: ai-integration-phase.md — #2529 MAJOR 1 (round 10): the workflow's pre-existing inline response-language directive is rewritten IN PLACE so the sentence names inter-tool NARRATION explicitly — narration between tool calls, status updates, progress notes, findings — instead of only "questions, prompts, and explanations". That older wording is the defect #2529 reports (it leaves the running commentary between tool calls in English while the answers around it are translated), and `scripts/lint-response-language-coverage.cjs` had been certifying it as coverage, so the gate legitimised the bug. +87 bytes, prose only: no step, gate, tool invocation, or subagent dispatch shape changed. FILE delta and LOADED-CONTEXT delta are both +87 here, and that identity is the point — the directive was deliberately NOT converted to an `@`-reference, because an `@`-import in this repo is EAGER (ADR-1610 Decision point 4; docs/ARCHITECTURE.md: moving prose into a file that is still eagerly `@`-imported "shrinks the measured file without shrinking loaded context"), so the conversion would have bought a smaller measured file at a cost of 1,692 bytes of loaded context per workflow — the 63-byte import line plus the 1,629-byte reference — against the 87 bytes this inline directive costs, a net +1,605. Stated plainly because the gates cannot state it: the repo's size gates — the tier hard caps in `tests/workflow-size-budget.test.cjs` and this size ratchet — measure the FILE, not the transitive inline, so an `@`-reference conversion would have READ as a smaller change to every gate in the repo while costing 1,605 bytes more loaded context per invocation. Re-homed in round 16: the fragment that carried this sentence (`3423-required-reading.json`) was retired on `next` by |
||
|
Tom Boucher
|
10593e2cf2 |
fix(#3959): give review-prompt plans citable path anchors (#4170)
* test(#3959): plan copies carry source names, prompt carries path anchors Regression coverage: budget copies named gsd-review-plan-<plan-id>.md (not a bare padded index), no bare-index copy survives, and the Plans to Review template instructs a per-plan repo-relative #### path header. Also corrects the stale -00 assertion to the source-named copy. * fix(#3959): give review-prompt plans citable path anchors The Plans to Review template now instructs a per-plan repo-relative #### path header, and the budget copies are named gsd-review-plan-<plan-id>.md instead of a bare padded index, restoring provenance for prompt-budget's per-plan headers while keeping the trim glob intact. Emitted-Drift-Ack-Growth: review.md — per-plan path-header instruction in Plans to Review + provenance-named copy loop (#3959) * chore(#3959): backfill changeset pr number --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
e242d85c92 |
fix(#4109): rewrap unquoted DISPATCH_SLUGS-shaped consumption to survive zsh (#4116)
* test(#4109): add zsh regression coverage for gate-check and invoke_reviewers dispatch Extends tests/review-plan-coverage-manifest.test.cjs to cover the two DISPATCH_SLUGS consumption sites the existing #3301 coverage-check tests don't reach: the write_reviews gate-check block (ALL_LANES_SKIPPED / TOTAL_LANE_FAILURE counting) and invoke_reviewers' dispatch + join loops. Both extract the real shipped bash from review.md and run it under bash and zsh, same convention as the existing coverage-check rows. Expected RED under zsh pre-fix, GREEN post-fix. * fix(#4109): rewrap unquoted DISPATCH_SLUGS-shaped consumption to survive zsh Bash word-splits an unquoted scalar on IFS by default; zsh does not (no setopt SH_WORD_SPLIT anywhere in these files), so a scalar accumulated from multiple space-separated tokens and then consumed via bare `for x in $VAR` collapses onto one bogus iteration under zsh whenever it holds 2+ tokens. Fixes all four review.md sites reported by #4108/#4109 (invoke_reviewers dispatch + join loops, the gate-check block, and the #3301 coverage-check block), plus the identical pattern found by a repo-wide sweep in complete-milestone.md, code-review.md, pr-branch.md, sync-skills.md, and execute-phase/steps/per-plan-worktree-gate.md. Each site is rewrapped in unquoted command substitution (`$(printf '%s' "$VAR")`), which re-splits identically under both shells regardless of SH_WORD_SPLIT — the same mechanism that already made the accumulator-building loops in these files shell-safe. * test(#4109): warn loudly when the zsh probe fails instead of skipping silently detectShells() drops the zsh test lane whenever a live zsh probe fails (e.g. on a CI runner without zsh installed), and a dropped lane reads identically to a passing one in the suite's own output — exactly the blind spot that let #4109's bug class ship undetected. Both copies of this helper (this file and its byte-identical duplicate in review-build-prompt-optional-sections.test.cjs) now emit a greppable console.warn when the probe fails, so a run without zsh reads as "zsh coverage unknown" rather than "all lanes green". * ci(#4109): gate workflows/*.md's embedded bash blocks in lint:ci Adds scripts/lint-workflow-shellcheck.cjs, wired into npm run lint:ci, so this bug class can't land undetected a third time. Two independent checks run over every ```bash block in gsd-core/workflows/**/*.md: - ShellCheck (new `shellcheck` devDependency, downloads and caches the real koalaman/shellcheck binary) catches the general unquoted-expansion/ word-splitting family in argument position. 212 pre-existing findings across the tree are absorbed into scripts/lint-workflow-shellcheck-baseline.json (matched on {file, code, message}, not line number, so unrelated edits elsewhere in a file can't spuriously un-baseline anything) — fixing all of them is out of scope for this issue; only NEW findings fail the build. - A custom structural check specifically for #4109's own shape: ShellCheck does not flag a bare `for x in $VAR` word-list — it treats that as an intentional idiom under any ruleset (confirmed empirically). This check does, and gates the build on any occurrence, with zero tolerance (no baseline) since every known site was already swept and fixed on this branch. * test(#4109): update pr-branch cherry-pick loop test anchor for the zsh fix extractPickLoop()'s PICK_LOOP_MARKER located the create_pr_branch cherry-pick loop by the literal substring "for HASH in $INCLUDED_COMMITS", which no longer appears verbatim after this issue's fix rewrapped that loop in unquoted command substitution. Updates the anchor (and its error message, now derived from the same constant instead of duplicating stale text) to the new literal form. No behavior change to the extraction logic itself. Emitted-Drift-Ack-Growth: review.md — #4109's fix adds explanatory comments at 4 sites; net code is functionally equivalent, comment expansion grows the file Emitted-Drift-Ack-Growth: complete-milestone.md — #4109's fix adds explanatory comments at 3 sites documenting the zsh word-splitting divergence Emitted-Drift-Ack-Growth: code-review.md — #4109's fix adds an explanatory comment documenting the zsh word-splitting divergence Emitted-Drift-Ack-Growth: pr-branch.md — #4109's fix adds explanatory comments at 3 sites (TRANSIENT_DIRS, INCLUDED_COMMITS, FILTER_PATHS) Emitted-Drift-Ack-Growth: sync-skills.md — #4109's fix adds explanatory comments at 2 sites (CREATE_LIST/UPDATE_LIST, REMOVE_LIST) * test(#4109): cover lint-workflow-shellcheck parsers + add timeout Adds tests/lint-workflow-shellcheck.test.cjs covering the hand-rolled parser/logic functions exported by scripts/lint-workflow-shellcheck.cjs (stripCommandSubstitutions, stripShellComments, substitutePlaceholders, extractForLoops, findBareForLoopSplits, findingKey, partitionAgainstBaseline) that shipped with zero coverage — CLAUDE.md requires at least one fast-check property test for parsers, included here (bare/braced forms always flagged naming the variable; quoted/substituted/literal forms never are). Also bounds runShellcheck's subprocess with a 60s timeout: the `shellcheck` npm package's own API has no timeout option and internally blocks on a synchronous spawnSync, so this reimplements the binary resolve/download step via the package's own exported config/download and calls spawnSync directly with a native timeout. And guards the CLI entry point with `if (require.main === module)`, matching this repo's sibling dual-purpose lint scripts — without it, requiring the module for its exported functions (as the new test file does) also triggered a live ShellCheck run as a side effect. * docs(#4109): add changeset for the zsh word-splitting fix * chore(#4109): backfill changeset PR number (#4116) --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
996196fe02 |
fix(#4099): stop word-splitting an unquoted PLAN_IDS string (bash vs zsh diverge) (#4102)
* fix(review): stop word-splitting an unquoted PLAN_IDS string (bash vs zsh diverge) #3301's plan-coverage-manifest block built a space-separated PLAN_IDS accumulator and re-split it via unquoted `for x in $PLAN_IDS`. Bash word-splits an unquoted scalar on IFS by default; zsh does not, so every plan id collapsed onto one iteration under zsh whenever a review had 2+ plans, undercounting the manifest and merging bullets onto one line. This is exactly what was failing origin/next's CI on every push. Derive PLAN_COUNT and the bullet list directly from the same *-PLAN.md glob loop used for copying — glob iteration, unlike re-splitting a stored string, is identical under bash and zsh, so it never needs word-splitting. Closes #4099 Emitted-Drift-Ack-Growth: review.md — fix comment documents the bash/zsh word-splitting divergence (gsd-core#4099); net code is smaller, comment expansion grows the file * docs(#4099): add changeset for the plan-coverage-manifest zsh fix * docs(#4099): fix changeset body to match the bold-dash-explanation format Code review (standards axis) flagged the changeset body as not matching CONTRIBUTING.md's canonical `**<Bold>** — <explanation>.` format — the bold clause had its own trailing period instead of being followed directly by the em-dash separator. * chore(#4099): backfill changeset PR number (#4102) --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
6eea00b707 |
enhance(#3301): tell reviewers the plan ids and total count, grade coverage (#4084)
* test(#3301): add failing-first plan coverage manifest tests Failing-first regression tests for the plan-id manifest, the updated Review Instructions, and the mechanical per-reviewer coverage check, ahead of the review.md implementation. RED baseline before the fix lands. * test(#3301): raise allow-test-rule-refs unverified ceiling for new marker Adding tests/review-plan-coverage-manifest.test.cjs's source-text-is-the-product marker grows the unverified-exemption pool by one (282 -> 283), the same documented growth path scripts/lint-allow-test-rule-refs.cjs's own failure output names. Confirmed clean via 'npm run lint:allow-test-rule-refs' locally. * feat(#3301): tell reviewers the plan ids and total count, grade coverage build_prompt now derives a plan-id manifest from each *-PLAN.md filename (stripping the -PLAN.md suffix) and appends it, with the total plan count, to both gsd-review-instructions.md and gsd-review-prompt.md. The Review Instructions prose requires one heading-verbatim section per id before any cross-plan or overall-risk content. write_reviews grades each dispatched lane's real (non-stub, non-empty) review against that same manifest and records an optional plan_coverage: frontmatter block, present only when a lane is incomplete. The match escapes regex metacharacters in the id and excludes a preceding/trailing hyphen or word character as a boundary, closing the two traps named in the issue (a decimal phase like 12.6 satisfied by 12X6-01; a threat id like T-04-07 registering as coverage of plan 04-07). CodeRabbit is exempt, since it never receives the source-grounding prompt carrying the manifest. This closes the gap where a review that silently covers only some plans in a multi-plan phase is indistinguishable from one that covers all of them. * docs(#3301): add changeset fragment * test(#3301): use t.after() instead of try/finally for cleanup CONTRIBUTING.md bans try/finally inside test bodies. Code review caught this in the new coverage-manifest test file; switch every fixture-cleanup site to the approved t.after() pattern. * test: use t.after() instead of try/finally in #3300's build_prompt tests Pre-existing try/finally-for-cleanup pattern in this file (landed for #3300) violates CONTRIBUTING.md's explicit ban on try/finally inside test bodies. Surfaced incidentally while reviewing #3301's diff, which cites this file as its extraction-pattern precedent; fixed inline per the no-defer rule rather than deferred to a separate PR. * test(#3301): anchor coverage-check extraction on the fence line, not prose `.plans-manifest.md` also appears in write_reviews' own prose ahead of the ```bash fence, so indexOf found that occurrence first and the backward-walk-to-fence-open landed on the earlier, unrelated gate-check block instead. gsd-test caught this: coverage-check tests expecting a real verdict got null, because the wrong block ran and never writes .plan-coverage-<slug>.json. Anchor on the fence-only bash assignment line instead. Emitted-Drift-Ack-Growth: review.md — #3301 adds the plan-coverage manifest and mechanical coverage check to build_prompt/write_reviews. * fix(#3301): route id escaping through the canonical pattern seam ADR-3212 (epic #3212) consolidated ~44 hand-rolled regex-escape copies into one owner, src/pattern.cts's escapeRegex, specifically to stop this exact class of duplication. My coverage-check node -e script hand-rolled the identical metachar-escape regex — invisible to eslint-rules/no-adhoc-regex-escape.cjs only because it lives inside a workflow markdown file, not a .cts/.cjs source file the shape-matching guard scans. Require the compiled seam (gsd-core/bin/lib/pattern.cjs) instead, matching the established node -e-requires-a-compiled-lib idiom already used elsewhere in this workflow (code-review.md's code-review-flags.cjs/code-review-depth.cjs calls). Verified both named traps from the issue still resolve correctly under escapeRegex's RegExp.escape-backed implementation, which differs in escaped-text shape (hex-escapes hyphens/leading chars) but not match result. * test(#3301): run coverage-check block with cwd at the repo root The block's node -e now requires ./gsd-core/bin/lib/pattern.cjs, a path relative to the repo root (correct for production, which always runs from there). The test harness ran it with cwd at the fixture's own temp dir instead, so the require failed. Add an optional cwd param to runScript (default: root, unchanged for the plan-copy-block tests) and pass the real repo root for every coverage-check call site. Manually verified end-to-end before spending another remote run: the extracted block now produces the expected {complete:true} verdict. * docs(#3301): backfill changeset pr number (pr:0 -> pr:4084) --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
929e02cb2c |
enhance(#3885): no silent swallow, and no verdict manufactured from dropped data (#3925)
* test(#3885): failing-first coverage for the depth bound and the manufactured wave verdict ADR-3473 §8.5 says a swallowed failure may not become an authoritative-looking answer. Three families do exactly that today; this commit pins each one RED. Measured on this tree, 2026-08-27: intel query, .planning/intel/file-roles.json nested 12000 deep -> exit 1, "Error: Maximum call stack size exceeded" searchJsonEntries / matchesInValue carry no depth parameter at all. The MAX_JSON_SEARCH_DEPTH = 48 bound existed in the retired SDK lineage (sdk/src/query/intel.ts at 11918dcc3^) and the surviving .cts lineage never received it. same fixture nested 48 and 49 deep -> both return total=1 at exit 0, truncated=undefined Nothing distinguishes "searched to the bottom" from "stopped looking". query phase-plan-index, a plan whose depends_on names an unresolvable token -> warnings: ["Plan 03-02: declared wave: 2 but depends_on DAG places it in wave 1"] The token is never mentioned. computeDependencyLevels drops the edge with `if (!resolvedDep) continue;`, every plan becomes a root, and the tool then reports the author's correct wave: as the thing that is wrong. countPhasePlansAndSummaries with fs.readdirSync throwing EACCES -> hasContext:false, indistinguishable from a phase that simply has no CONTEXT.md. context_read_error is undefined. The shapes these tests assert against, chosen here so the implementation has a target rather than inventing one later: `truncated: boolean` on the intel query result, `unresolved: Array<{plan, token}>` from computeDependencyLevels, and `context_read_error: string | null` per analyzed phase. Deliberately green, and they must stay that way — each stops the fix from over-firing: depth 48 is found and NOT flagged truncated (the ceiling is inclusive) a shallow miss reports no truncation (noise control, N1) 10,000 siblings at depth 2 are unaffected (the bound is DEPTH, N2) a genuine wave: mismatch on a fully-resolved DAG still warns (N3) a genuinely missing directory is absent, not an error the emitted depends_on display mapping still passes an unresolved token through verbatim — already pinned by the existing #3785 test, so no duplicate was added T31 asserts at the consumer's output per ADR-3180 Decision 4(b): it runs the real CLI and reads the emitted JSON, because a unit assertion on computeDependencyLevels would have passed throughout #3427's life. Design: .gsd/phase/feat-3885-no-silent-swallow/40-design.md Test matrix: .gsd/phase/feat-3885-no-silent-swallow/50-test-matrix.md Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * enhance(#3885): no silent swallow, and no verdict manufactured from dropped data Implements ADR-3473 §8.5. A failure or a gap in the input stops being absorbed into an output that reads as authoritative. The recursion bound, restored but NOT verbatim (src/intel.cts) MAX_JSON_SEARCH_DEPTH = 48 is threaded through searchJsonEntries and matchesInValue, which carried no depth parameter at all. The bound existed in the retired SDK lineage (sdk/src/query/intel.ts at 11918dcc3^) and the surviving .cts lineage never received it — §8.3's "a consolidation may not delete an invariant along with the surface that held it", demonstrated. Measured before: a .planning/intel file nested 12000 deep exits 1 with "Error: Maximum call stack size exceeded". Reachable from a project document. The original returned a bare `false` at the ceiling. Restoring that verbatim would trade a crash for a silent "no match" when the truth is "I stopped looking" — the same class this epic exists to close, and ADR-3473 Decision 4 forbids it. So the bound carries a truncation signal: nesting 47 -> found, truncated false nesting 48 -> found, truncated false (the ceiling is inclusive) nesting 49 -> not found, truncated TRUE nesting 12000 -> exit 0, truncated TRUE, no RangeError A shallow document that simply has no match reports truncated FALSE — the flag means "I stopped early", never "I found nothing", or it would be noise. The bound is on DEPTH: 10,000 siblings at depth 2 are unaffected. The dropped edge is named, and stops being blamed on the author (src/phase.cts) computeDependencyLevels dropped every unresolvable depends_on token with a bare `continue`. Each drop makes a plan a root, so the whole phase collapses to wave 1 — and cmdPhasePlanIndex then reported the author's CORRECT wave: as the thing that was wrong. Before: warnings: ["Plan 03-02: declared wave: 2 but depends_on DAG places it in wave 1"] After: warnings: ["Plan 03-02: depends_on token \"nonexistent-token-3427\" does not resolve to any plan in this phase — edge dropped, wave placement for this plan may be unreliable"] The suppression is PER PLAN, never blanket: a plan with a fully-resolved DAG and a genuinely wrong wave: still gets the mismatch warning. resolveDependencyId stays two-tier — the shortFormToId third tier is §8.3/Phase 6's rule and is deliberately not built here. The emitted depends_on display mapping still passes an unresolved token through verbatim (#3785). No artifact from failed inputs (gsd-core/workflows/review.md, #3352) A failed lane leaves no result file, so "every lane failed" is exactly "the aggregate JSONL has zero lines" — the gate condition already existed as a byproduct. REVIEWS.md is no longer written in that case, and the commit step is skipped with it. A budget-SKIPPED lane also leaves no file and is NOT counted as a failure. Per-lane output and non-empty .err are preserved to .review-diagnostics/ before `rm -rf "{run_dir}"` destroys the only record that the lanes failed at all; the commit step names one file, never a glob, so the diagnostics are not swept in. Unreadable is not absent (roadmap.cts, gap-checker.cts, init.cts x2) Four callers collapsed an EACCES on a phase directory into [] and reported hasContext:false — byte-identical to a phase that simply has no CONTEXT.md. Each now names the directory it could not read. A genuinely missing directory stays absent rather than becoming an error, which is what keeps the fix from over-firing. Fatal errno folded into a retry set: audited, no defect found Reported as a verified negative rather than padded with a change. withPlanningLock was fixed by #1884/PR #3472; acquireStateLock by #3776; atomicRenameWithRetry and estimate-cli's renameWithRetry are correct by construction — bounded set {EPERM,EBUSY,EACCES}, bounded attempts, and they return or rethrow the final error rather than swallowing it. estimate-cli's sole caller surfaces that rethrow as write_error in its JSON output. Manufacturing a diff to make the checkbox look worked-on is the Goodhart outcome Decision 6 exists to prevent. Disclosed: R46 (the commit step names one file, never a glob) is a real regression guard but is NOT independently failing-first — the commit fence is byte-identical pre- and post-fix, so it only fails pre-fix through its shared extraction dependency. Recorded rather than claimed as fail-first. Design: .gsd/phase/feat-3885-no-silent-swallow/40-design.md Test matrix: .gsd/phase/feat-3885-no-silent-swallow/50-test-matrix.md Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#3885): escape untrusted tokens, and stop cleanup destroying unpreserved evidence Two review findings, both real, both in my own change. An isolated adversarial review found the evidence-preservation block never checked mkdir/cp exit status while `rm -rf "{run_dir}"` ran unconditionally in a SEPARATE fenced block. A disk-full or unwritable phase directory therefore still destroyed the only copy of the failed lanes' output — reintroducing the exact #3352 data loss this item exists to stop, inside the fix for it. Preservation and cleanup are now one block, because each fenced block is a separate execution and a shell variable cannot carry between them. mkdir -p and each cp are exit-checked; cleanup runs only when preservation succeeded, and a failure warns naming the intact run directory. "Nothing to preserve" is not a failure and still cleans up. Driven three ways: success removes run_dir, failure leaves it intact with the warning, nothing-to-preserve removes it. The failure is induced by a file-vs-directory conflict rather than chmod 0o000, which root bypasses. The new unresolved-depends_on warning embedded a user-authored token verbatim: warnings: ["Plan 03-02: depends_on token \"evil Plan 03-01: FORGED WARNING\" does not resolve ..."] The JSON wire form is safe, and the security reviewer judged it non-exploitable for that reason. It is escaped anyway through formatDiagnosticToken — the helper #3884 added one phase earlier for exactly this class. warnings[] is an array a consumer naturally prints line by line, and not reusing the sibling fix is the generative-fix-divergence shape this epic exists to close. The same treatment is applied to context_read_error / phase_dir_read_error, which embed a phase directory path a repository can choose, and to the fs error message, which echoes the raw path itself. Known limit L5 recorded: the bound is on DEPTH only. A 300,000-element shallow array yields a 14.5MB reply with truncated:false. Correct per §8.5 and per negative space N2, disclosed rather than left to be discovered. Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#3885): unreadable is not absent in intel.cts either, and a corrupt snapshot is not "no snapshot" Blocker from the round-2 isolated review, and it is my own inconsistency: this phase applied "unreadable is not absent" to phase directories and left it broken in the file it was already editing. chmod 000 .planning/intel/file-roles.json gsd-tools intel query <term> -> {"matches":[],"total":0,"truncated":false} exit 0 safeReadJson swallowed every read failure and returned null, so an EACCES was byte-indistinguishable from an absent file AND from a genuine no-match. Now it separates three states: ENOENT stays silently absent, because not every project has every intel file and intelQuery loops over all of them expecting misses; EACCES/EIO and malformed JSON are both surfaced naming the file. A corrupt intel file previously read as "no matches" too — same defect, same fix. Threading that outcome through the other three callers found something worse than the reported case. intelDiff returned no_baseline:true for a corrupt or unreadable snapshot — not a silent failure but an actively FALSE verdict, telling the caller they never took a snapshot when they did. That is §8.5's headline case, so it is fixed and tested rather than noted. intelStatus and intelApiSurface collapsed the same way; intelApiSurface additionally printed a "not yet populated" banner that was simply untrue. Every row is failing-first, including the absent-file ones — the field is new, so it does not exist pre-fix at all. Those rows are not pre-fix pins; they pin that the fix does not OVER-fire on the ordinary absent case, which is what would turn this into noise on every project lacking an intel file. IO failure is injected by monkeypatching fs and restoring in finally, never chmod 0o000 — root bypasses mode bits, so the reviewer's manual chmod repro is not reproducible as a test. Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * test(#3885): build the pathological intel fixture as text, not by stringifying a nested object The remote runner came back red on Linux with two failures, both T4: deeplyNestedIntelDoesNotOverflowTheStack, while the same test passed on macOS. The product was never at fault. writeNestedFixture(12000) built a 12,000-deep JavaScript OBJECT and then JSON.stringify'd it. JSON.stringify recurses once per level, so it overflowed the TEST PROCESS's stack — the error was thrown before the CLI was ever spawned. Linux's container stack is smaller than macOS's, which is the whole of the platform difference. Measured, with the same document built as JSON TEXT so nothing in the building process recurses: depth=100 rc=0 truncated=true depth=5000 rc=0 truncated=true depth=12000 rc=0 truncated=true depth=60000 rc=0 truncated=true V8 parses this shape iteratively; only stringify recurses. The bound works at every depth tried. The fixture is now built by string concatenation. That is also the more faithful input — a real deeply nested JSON document on disk is exactly what the bound guards, where a stringified object was only ever a way to produce one. The depth stays 12000. Lowering it would have made the test pass by weakening it to accommodate a fixture bug, and 12000 is a legitimate pathological input the product handles. T4 remains a genuine fail-first: rebuilt against the parent of the commit that added the bound, the string-built depth-12000 fixture still drives the CLI to rc=1 with "Error: Maximum call stack size exceeded". A comment records why the fixture is text, so it is not "simplified" back into a macOS-green / Linux-red test. Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#3885): backfill the changeset PR number Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * test(#3885): normalize path separators before splicing into the workflow's bash CI red on one lane — test (windows-latest, 24, shard 3/3). macOS, Linux and the remote runner were all green. AssertionError: commit must name the single REVIEWS.md file; got: --files C:UsersRUNNER~1AppDataLocalTempgsd-3352-phasedir-mOKmuy/03-REVIEWS.md Every backslash in C:\Users\RUNNER~1\AppData\Local\Temp\... was eaten. The harness spliced an OS-native temp path into the extracted bash, and bash consumes \U, \A, \L and \T as escapes on an unquoted expansion. The same loss broke RUN_DIR, so "rm -rf" targeted a path that never existed and the run directory survived — which is the other two assertions. This is a fixture defect, not a product one, and that was checked rather than assumed. In production the phase directory is toPosixPath-normalized at every call site that serializes it (bin/lib/init.cjs:951, 1381, 1461, 1529, 1595), and the run directory is created by "mktemp -d" running inside the bash block itself (gsd-core/workflows/review.md:163), which emits POSIX-style output even under Git-Bash on Windows. Neither ever carries a backslash where the workflow reads it. The file's pre-existing #3034 harness splices raw native paths too, but only ever inside double-quoted assignments, so it never tripped this — my new harness followed that convention faithfully into the one place where it does not hold. Both now splice through toPosixPath from shell-command-projection, the established seam, which is a no-op on POSIX and mirrors what production does. No assertion was weakened. "commit must name the single REVIEWS.md file" and "the run dir must still be destroyed" still assert exactly that; only how the fixture supplies its path changed. Nothing is skipped on Windows — a t.skip() here would have hidden the question of whether the exposure was real, which is the question that mattered. Driven both ways: a synthetic C:\Users\RUNNER~1\... input reproduces the exact CI string when unfixed and yields C:/Users/RUNNER~1/... when fixed; a POSIX input produces a byte-identical shape, proving the normalization is idempotent. Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * test(#3885): stop the harness making the deleted run dir its own cwd Windows shard 3/3 stayed red after the separator fix, on two assertions the separator fix never touched: AssertionError: the run dir must still be destroyed AssertionError: nothing to preserve is not a failure — run dir must still be removed The separators were a real bug and fixing them fixed the --files assertion. They were not this bug, and two CI cycles went into the wrong axis before I stopped converting path forms and looked at what the harness actually does. runWriteReviewsFlow passed cwd: runDir to runHook, so the child bash process's working directory WAS the directory the block under test then removes with rm -rf "$RUN_DIR". POSIX allows a process to delete its own cwd — verified locally, cd "$d"; rm -rf "$d" removes it cleanly — and Windows does not: a live process's working directory cannot be removed. So on Windows the directory survived and both assertions failed, on macOS and Linux it vanished and they passed. Nothing to do with slashes. Harness-only. Production never cd's into the run directory; every reference is by absolute path, and RUN_DIR is created by mktemp -d inside the bash block itself (gsd-core/workflows/review.md:165) rather than injected. review.md is unchanged. Fix: the child now runs with its cwd in an unrelated temp directory that the block under test never deletes. Neither assertion was weakened, and nothing is skipped on Windows — the tests in this file carry no platform guard and run there unconditionally, which is how this surfaced at all. Honest limit: the Windows failure mode cannot be reproduced on macOS, because POSIX permits the very thing Windows refuses. The diagnosis is grounded in that documented divergence and in the fact that only the Windows lane failed, but the green outcome on windows-latest is unverified until CI runs it. Refs #3885 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: sim <sim@local> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
fb2d122d7f |
feat(#3841): assert gsd-tools identity on every state-mutating verb (#3848)
* feat(#3841): assert gsd-tools identity before any state-mutating verb only this package publishes. The path-based branches — a project-local install, a runtime config directory — had no such guarantee; they trusted their configured location. This closes them. Mechanism: once resolution finishes, and before any verb runs, the preamble probes the tool it picked with `runtime-identity --raw` and matches the answer with a shell `case` pattern ANCHORED to the start of the compact payload (`{"packageName":"@opengsd/gsd-core"`). An unanchored substring match accepts the decoy `{"packageName":"get-shit-done-cc","note":"@opengsd/gsd-core"}`, which any colliding package could publish. The outcome is exported as the two-valued `GSD_IDENTITY_STATUS` (`ok`/`unverified`), so the gate is asserted on a VALUE rather than on warning prose. Rollout is warn-then-fail per the #3146 ruling: `unverified` prints one line naming BOTH causes and continues, because `no_identity_verb` cannot tell a foreign package from an `@opengsd/gsd-core` older than the verb, and at rollout the old-version case is the common one. The blocker was byte budget, not design. The preamble is inlined into 112 shipped files and several sat within single-digit bytes of frozen ceilings (`gsd-verifier.md` 16 bytes, `gsd-executor.md` 33, `execute-phase.md` 234); a first attempt broke five of them. What made room was collapsing the resolver's twenty near-identical `elif [ -f … ]` arms into one candidate-list helper (`_gsd_at`), which buys far more than the assertion costs. The preamble is now 2,624 bytes against 4,500 — a net 1,876 bytes SMALLER per inlined file, so every capped file moved away from its ceiling rather than toward it. No cap raised, no size-budget exception added, no override token emitted. Resolution order, every runtime-home probe, the `unset -f gsd_run` re-source fix, the fail-closed `exit 1`, and the `CLAUDE_ENV_FILE` persistence are all preserved byte-for-byte in substring terms; the snippet still begins with `_GSD_SHIM_NAME=` and still ends with `fi`, which the parity extractors anchor on. `gsd-core/references/gsd-run-resolver.md` is re-synced byte-equal. Also fixes two stale claims found in passing: CONTEXT.md and FEATURES.md both described an `[ -x ]` guard as the load-bearing re-source defense. That guard was tried and REMOVED in #3831 — it rejected the bare function name, fell through every branch, and hit `exit 1`, which kills a sourced caller's shell. `unset -f gsd_run` is the actual mechanism. Refs #3841 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#3841): pair the anchor's brace by requiring a closed identity payload The matrix went red on `tests/new-project-mvp-prompt.test.cjs` — "new-project.md has unbalanced braces: net depth 2" — plus a knock-on report from its parent `bug #1516` describe, which is the same failure counted once at the child and once at the block. Root cause: that guard (:182-189, mirroring #3784 bd53925f) walks characters and increments on `{`, decrements on `}`, with no awareness of shell quoting. It scans `new-project.md` PLUS every `new-project/steps/*.md`, and both `new-project.md` and `steps/auto-mode-config.md` carry one inlined preamble copy — hence net 2 from a snippet that was off by exactly one. The unpaired brace was the `{` inside the single-quoted `case` pattern of the identity anchor, which is correct shell and invisible to a text scanner. Fix in the snippet, not the guard. The pattern now anchors at BOTH ends: `'{"packageName":"@opengsd/gsd-core"'*'}'`. That balances 51/51 with a brace that does real work rather than a cosmetic pair — a truncated payload whose prefix matches now fails too, where before it verified. Safe for any future additive field: a JSON object's own closing brace is always the last character, whatever type the last value has, which is pinned by two negative-space tests (a nested object and an array-valued last key must both still verify). Cost: +3 bytes, against the 1,873 the resolver fold already gave back. The alternative considered and rejected was dropping the literal `{` for a `?` glob. It balances too, but weakens the anchor from "must be an opening brace" to "must be any one character", and the anchor is the entire point. Two guards added so this cannot recur silently: - runtime-launcher-parity (F0) pins brace balance at the SNIPPET, so the next edit to that pattern fails on the file it broke instead of surfacing three files downstream in a test whose name mentions neither the launcher nor this issue. It also asserts depth never goes negative, since a `}` preceding its `{` nets to zero while being unbalanced at every prefix. - runtime-identity gains behavioral truncated-payload and trailing-garbage fixtures, so the added `}` is proven load-bearing rather than merely present. Verified: snippet 51/51 braces; new-project combined net depth 0; the seven other preamble-bearing files with nonzero depth are unchanged from merged next (their own prose, not the preamble, and not in any guard's scan set); all 112 inlined copies and the resolver reference re-synced byte-equal; sync:launcher idempotent on the second run. Refs #3841 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#3841): backfill changeset PR number Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: sim <sim@local> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
63abcface9 |
feat(#3146): resolve gsd_run so workflows cannot reach a foreign gsd-tools (#3831)
* feat(#3146): resolve gsd_run so workflows cannot reach a foreign gsd-tools The predecessor package get-shit-done-cc publishes a colliding gsd-tools bin whose phases.clear DELETES where this package's ARCHIVES, and both print success-shaped output against a gitignored .planning/ -- which is how #3129 cost a user 43 phase directories with no error and nothing recoverable from git. The launcher's PATH branch now resolves gsd_run, published only by this package and self-locating via its own symlink chain to the sibling shim, instead of the colliding gsd-tools. A foreign handler becomes unreachable from PATH, and when no gsd_run is reachable the resolver fails closed rather than falling back -- that fallback was the vulnerability. This is smaller than the branch it replaces, which matters: the preamble is inlined into 113 shipped files and agents/gsd-verifier.md sits 2 bytes under a red-line size cap. unset -f gsd_run leads the preamble so a re-source is idempotent. Without it, command -v finds the shell function, returns a bare name, and the resolver falls through to an exit 1 that kills a sourced caller's shell. Adds gsd-tools runtime-identity, a manual diagnostic reporting this runtime's package coordinates over the baked package-identity (#498) and readHostVersion, with a strict total classifier: only a JSON object with an exact packageName verifies, since JSON.parse admits 0/"str"/[]/null/true. An inlined identity assertion was built and reviewed first, then withdrawn -- it breaks five frozen size ceilings and no assertion fits in 2 bytes. Closes #3146 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#3146): stop sync:launcher relocating a deliberate preamble placement Pre-existing defect, surfaced by this PR because sync is a no-op unless the snippet content actually changes. transformFile inserts the preamble into the first block that CALLS gsd_run, but gsd-core/workflows/explore.md deliberately places it in a bootstrap-only block that DEFINES gsd_run without calling it -- its own comment explains why: declining the research offer must not leave Step 5's commit call unbootstrapped. Stripping empties that block of calls, so the preamble migrated forward and broke the define-before-use invariant tests/explore-command.test.cjs pins. Reproduced on a pristine origin/next checkout with the base snippet and base file, so this was not introduced here. The insertion target now honours a block that already carried the preamble, falling back to the first calling block for files that have none yet. Adds a behavioral regression test over a two-block fixture. Also updates three runtime-launcher-parity tests that pinned the removed PATH fallback to gsd-tools. Their intent is preserved -- the PATH stub is renamed gsd_run so it is reachable by the new resolver, and the RUNTIME_DIR-wins test still asserts the stub is never invoked. Fixture shebangs move to an absolute /bin/sh, because the fixture PATH is deliberately restricted and #!/usr/bin/env sh could not resolve. Refs #3146 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#3146): backfill changeset PR number Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * docs(#3146): document the FEATURES.md section-numbering practice The monotonically increasing section number in docs/FEATURES.md is the most frequent merge-conflict source in this repo, and it has TWO conflict cells, not one: the ### N. heading and the hand-maintained table of contents. Two PRs adding differently numbered features still collide on the TOC, so renumbering alone does not make a branch safe. This branch alone was renumbered 165 -> 166 -> 167 -> 168 across successive rebases. Adds a CONTRIBUTING section stating the practice: allocate the number last, never pre-emptively renumber, take max+1 after a rebase and update the TOC in the same commit, and never renumber someone else's section. Fork contributors are told explicitly they may leave the number to a maintainer at merge rather than chasing the counter. Agents are told to lease the allocation and to include the file in their published touched set. Records the durable fix as planned rather than pretending it exists: FEATURES.md should be generated from per-feature fragments the way CHANGELOG.md is generated from .changeset/, and the way tests/emitted-drift-acks/ works (#2914). Also renumbers this branch's own section to 168, leaving 167 to the PR already in flight. Refs #3146 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: sim <sim@local> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
a2387a0545 |
feat(#3034): add opt-in parallel reviewer lanes (#3822)
* test(#3034): failing-first coverage for opt-in parallel reviewer lanes Executes the real invoke_reviewers dispatch block from review.md against a stubbed gsd_run seam rather than pattern-matching the workflow text, so the two properties that actually carry risk are observable: that every lane is joined before aggregation, and that concurrent lanes cannot tear a line in gsd-review-lane-results.jsonl. Concurrency is proven by a barrier fixture, not by elapsed time -- each stub lane blocks until all lanes have checked in, which can only complete if they overlap. Red against the current sequential dispatch, by design. Refs #3034 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * feat(#3034): add opt-in parallel reviewer lanes Reviewer lanes within one review pass inspect the same immutable plan snapshot and have no dependency on one another, but were dispatched strictly one at a time, so a multi-reviewer pass cost roughly the sum of its lanes. The serialization is a deliberate protection against provider rate limits, so it stays the default; review.parallel_lanes opts a project out of it. The loop body is hoisted into run_review_lane so the sequential and concurrent paths share one body -- two hand-synced dispatch bodies is the divergence class ADR-2782 spent a phase deleting. Each lane writes a slug-scoped result file, concatenated in selection order after the join: concurrent O_APPEND is atomic only below PIPE_BUF, and write_reviews parses that JSONL to render the models:/model_sources: frontmatter, so a torn line is a broken REVIEWS.md rather than a cosmetic log defect. Aggregating in selection order also keeps the artifact byte-identical between the two paths. The guard is strict equality on "true" and falls back to sequential when config-get fails -- the opposite polarity from the commit_docs guard, because failing open here fires the very requests the default prevents. Also corrects docs/COMMANDS.md and its four locale mirrors, which described --all as running every configured reviewer in parallel when dispatch was in fact sequential. Closes #3034 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#3034): de-duplicate dispatch slugs and scope lane locals Review finding (Standards axis): a slug repeated in SELECTED_REVIEWERS would put two concurrent background jobs on the same > -truncated per-lane result file. The shared-append form this replaced could not corrupt itself that way, so de-duplicating is what keeps the concurrent path no worse than the sequential one. Selection de-dupes today -- the roster is a Set and review.default_reviewers normalizes lowercase-unique -- but reachability analysis is not a contract, which is the same reason the roster derivation itself is guarded. Splitting once into DISPATCH_SLUGS also removes the duplicated tr-split the same review flagged: the dispatch and aggregation loops now share one list, which is what guarantees they walk the same slugs in the same order. A plain string accumulator rather than an array, because zsh and bash disagree on array indexing and this block runs under both. Also scopes run_review_lane's locals. Not a live fix -- each dispatched call already forks its own subshell -- but it makes the isolation a property of the function rather than of the dispatch mechanism happening to fork. Refs #3034 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * test(#3034): acknowledge review.md growth, drop spent 2295 ack The differential attribution gate reported review.md growing 4173 bytes (30712 -> 34885) with no live acknowledgment. Adds the per-PR fragment it asks for, naming only the one path it reported. Deleting tests/emitted-drift-acks/2295-resolved-model.json is required, not opportunistic. That fragment declared review.md and nothing else, and its ripple is already absorbed into the base, so it is spent -- it can no longer clear anything, which is why the gate still reported review.md as unacknowledged. It could not simply be left alone either: two ack sources may never name the same path, so it blocked this PR's fragment outright. CONTRIBUTING is explicit that a fragment whose last entry is removed gets deleted with it, because an empty fragment signals nothing while its presence reads as a live alarm. Refs #3034 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#3034): backfill changeset PR number Refs #3034 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: sim <sim@local> Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
cf15682d1c |
enhance(#3028): responsive Markdown separators instead of fixed-width rules (#3789)
* feat(#3028): responsive Markdown separators instead of fixed-width rules Stage banners, checkpoints, completion and error panels used fixed-width runs of box-drawing characters -- a 53-column heavy rule and a 62-column double-line box. Those runs are ordinary text to a Markdown-rendering host, so in a narrower pane they wrap and the border comes apart from the heading it framed. Shipped content now emits an ATX heading for a titled section and a blank-line-delimited --- for a break between sections, both of which adapt to the available width. The same convention is applied to the three code sites that built these strings at runtime: the UAT checkpoint renderer, the milestone-close audit report, and the TDD review checkpoint table. Removing the box also removes its only reason to exist -- the east-asian-width padding helpers that kept its right border aligned (checkpointBoxLine, displayWidth, isWideCodePoint, ZERO_WIDTH_MARK_RE, CHECKPOINT_BOX_WIDTH). RTL directional isolation is unchanged. The convention is specified in gsd-core/references/ui-brand.md and enforced across all shipped content by tests/responsive-separators.test.cjs. Refs #3028 * test(#3028): pin the heading form in checkpoint and audit-report assertions These suites asserted the exact box borders and the 62-column padded banner interior. With the box gone they assert the ### heading form, the --- break and the bolded instruction line, and each now carries a positive assertion that no box character remains -- which is what pins the fix rather than merely tolerating it. Language coverage is converted, not dropped: Japanese, Chinese, Korean, Hindi and Arabic all still assert their rendered banner, and the Arabic case still asserts the RTL directional isolates the box removal must not disturb. Adds a case for a banner longer than the old inner width, which previously produced a ragged border and now has none. Refs #3028 * chore(#3028): acknowledge execute-plan.md growth from the checkpoint display spec The checkpoint_protocol display spec described the drawn box; it now describes the heading, the --- break and the bolded action prompt, which costs 22 bytes (40111 -> 40133, 827 under the cap). Appended to the existing #3370 fragment rather than filed as a new one: a growth ack keys on the bare filename and #3370 already declares execute-plan.md, so a second source naming it would be a hard duplicate-key error. Same supersede-by-append route #3370 took for the spent #2652 fragment. Refs #3028 * docs(#3028): state the load-bearing half of the separator rule, and amend the zh-CN reference Review found three things. The rule as first written demanded a blank line above AND below every ---. Only the one above is load-bearing: it is what stops CommonMark reading the rule as a setext underline for the line above. The one below is cosmetic, because a thematic break is a leaf block. The rule now says that, with the reason, instead of asserting a stricter form the content does not keep. The zh-CN reference had received the mechanical box-to-heading swap but none of the prose behind it: it still claimed a 62-character checkpoint width and still listed --- among forbidden mixed banner styles, so it contradicted the convention it was translating. It now carries the separator section, the setext reasoning, the unconditional-vs-per-runtime rationale and a corrected anti-pattern list, in Chinese. The user guide asserted that a heading is not a degradation anywhere. That is an assertion, not a demonstration. It now says what was actually traded away in a plain terminal, points at the recorded rationale, and invites the report that would justify the capability flag instead. Refs #3028 * chore(#3028): backfill changeset PR number Refs #3028 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
1bf73d957b |
enhance(#2295): record the resolved model per reviewer in REVIEWS.md frontmatter (#3649)
* test(#2295): failing-first coverage for per-lane resolved-model recording * feat(#2295): record the resolved model per reviewer lane * docs(#2295): document the recorded reviewer model and its provenance * fix(#2295): refuse control characters in a recorded model value * test(#2295): correct watermark assertions for the widened mark shape * fix(#2295): anchor the role-manipulation injection pattern at a word boundary * feat(#2295): record the applied reasoning effort in the model value * chore(#2295): backfill changeset pr number * chore(#2295): restore em-dash in changeset body --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
011153c076 |
fix(#3300): guard build_prompt optional sections against nullglob (#3454)
* fix(#3300): guard build_prompt optional sections against nullglob * chore(#3300): backfill changeset pr number 3454 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
483083aea6 |
fix(#3194): verify source-grounded lane evidence from review output (#3436)
* fix(#3194): verify source-grounded lane evidence from review output * chore(#3194): fill changeset pr number --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
e7ce60fd21 |
fix(#3035): add kimi-code detection and flag to review workflow (#3115)
* fix(#3035): add kimi-code detection and flag to review workflow The kimi-code reviewer lane was declared in REVIEWER_LANES, documented in docs/COMMANDS.md, resolved via --kimi-code, and functional when reached — but review.md's detect_clis hardcoded 11 of 12 lanes (no kimi probe) and the flag-parse list omitted --kimi-code. /gsd:review --kimi-code could never reach SELECTED_REVIEWERS. Added command -v kimi detection and the --kimi-code flag to the review workflow's CLI detection and flag-parse steps. * chore(#3035): backfill changeset PR number 3115 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
589a9b29b0 |
fix(#2962): enable nullglob in for-glob shell blocks for zsh portability (#3087)
* fix(#2962): enable nullglob in for-glob shell blocks for zsh portability Workflow shell blocks are fenced bash but execute in the user's login shell (zsh on macOS). zsh's nomatch default aborts the WHOLE block on an unmatched glob in a for-list (not just skipping the command), silently bypassing every statement after it — including the verify-phase decision-coverage gate, whose optional *-CONTEXT.md lookup used the unsafe for-list form so the DECISION_RESULT= assignment on the next line never ran under zsh. Fix: prepend a portable nullglob shim to every bash block containing a for-glob loop: shopt -s nullglob 2>/dev/null; setopt NULL_GLOB 2>/dev/null Each command no-ops (stderr suppressed) in the shell that doesn't recognize it; the matching shell enables nullglob so an unmatched glob expands to nothing and the loop body is skipped cleanly. Verified locally: both zsh and bash now reach end-of-block (rc=0) on a no-match glob; bash matched-case behavior unchanged. 14 blocks across 7 files: verify-phase.md (4, incl. the decision-coverage gate), review.md, execute-phase.md, resume-project.md, complete-milestone.md, audit-milestone.md, gsd-integration-checker.md, gsd-plan-checker.md (3). Closes the zsh bypass of the #2770 fix. * chore(#2962): add changeset fragment * chore(#2962): backfill changeset PR number 3087 --------- Co-authored-by: sim <sim@local> |
||
|
Tom Boucher
|
ff4a57b78c |
chore(#1671): migrate the remaining 13 LARGE/XL workflows to the fragment model — Phase 6.3 (#3030)
* chore(#2994): fragmentize progress.md forensic audit onto the fragment model Extract the --forensic-gated forensic_audit step to workflows/progress/steps/forensic-audit.md behind a section marker, and repair progress.md's init line to forward --forensic so the atom is actually true in production rather than only under direct CLI tests. progress.md shrinks 32630 -> 27207 bytes. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize the four manifest-wired workflows new-project, quick, new-milestone and progress each already had a dedicated cmdInit* entry point but zero marked sections. Extract nine gated bodies to workflows/<wf>/steps/ behind section markers and repair each init line to forward its flags. Fold --full into the discuss/research/validate facts inside cmdInitQuick so the when= grammar never sees an OR, per the chunked-mode precedent. Fixes found while working, per the no-defer rule: - cmdInitProgress passed no phase info to buildSectionManifestField, so state:phase-mvp-mode was permanently false — an atom in the vocabulary whose fact could never be computed. - the quick init router folded flag tokens into the free-text description, which the new forwarding would have corrupted. - a #2508 dispatch note was nested inside quick.md's Agent(prompt=) fence, leaking orchestrator guidance into the subagent prompt. - progress.md had a 3-vs-4 backtick outer-fence imbalance. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize verify-work.md and admit state:ui-phase-active Wire cmdInitVerifyWork to buildSectionManifestField — it was a dedicated entry point that never emitted a manifest — and mark two sections. state:ui-phase-active folds (plan:pre hooks include an active ui step) OR (the phase dir holds a *-UI-SPEC.md) into one boolean in init.cts, so the grammar still sees a single operator-free atom. The inner Playwright-MCP check stays as prose inside the fragment: it is live session state and no init seam can precompute it. The MVP false-branch note is a real fallback, not redundant prose, so it sits outside the marker — gating it away would delete the text needed precisely when MVP mode is off. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(#2994): follow moved workflow content in drift guards Retarget every guard that asserted on content this branch moved into workflows/<wf>/steps/, mirroring 815b3d897. Each retargeted assertion was verified to still fail when its step file is blanked, so none was weakened into vacuity. Three assertions in verify-mvp-uat were genuinely red. Three more were worse than red — passing for the wrong reason: - quick-commit-boundary and worktree-cleanup anchored on indexOf('Step 5.6'), which matched a later cross-reference and sliced 16069 chars that coincidentally held the asserted substrings. Replaced with an expandWorkflowSections helper that splices step content back in place. - phase6-review-capabilities lost its end boundary and widened to EOF. - playwright-ui-verify matched 'UI' in an unrelated bullet and 'fall back' in a subagent-dispatch line after the real content moved. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize code-review and complete-milestone, admit three atoms Add dedicated cmdInitCodeReview and cmdInitCompleteMilestone entry points alongside the shared generic ones rather than modifying them — init.phase-op and init.manager carry a CRITICAL blast radius (179 dependents, 24 processes) and stay byte-identical for their other callers. Admit flag:--fix, state:fallow-enabled and state:git-create-tag, each with a consuming section and a fact its own entry point computes. Both sections had the resolver-in-body hazard: the fallow config-gate and the git.create_tag check each sat inside the very block being gated, so gating would have disabled the resolver that decides the gate. Both are hoisted into init and the bodies now consume the resolved fact. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(#2994): retarget code-review and milestone drift guards, fix two red tests Retarget guards that asserted on content moved into steps/, proving non-vacuity by blanking each step file and confirming failure. Also fixes two genuinely red tests found while working, per the no-defer rule: - workflow-fragments' frozen-vocabulary lock was missing state:ui-phase-active, so commit 7ef7f8336 shipped red. Lint and build both passed over it, which is why neither is sufficient verification. - code-review's quick.md capability-hook assertion carried a stale delimiter after the 18ff35d20 extraction. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize autonomous.md and admit state:plan-strategy-converge Five sections share one atom, the pattern plan-phase already uses for flag:--research-phase. The atom folds --converge OR --cross-ai into a single boolean in cmdInitAutonomous so the grammar stays operator-free. cmdInitAutonomous is additive; init.milestone-op, init.manager and init.phase-op are untouched and still consumed. The $PLAN_STRATEGY bash resolver is deliberately retained — ungated local-planning bullets still read it, so the init-side fact supplements it rather than replacing it. converge-fail-fast required splitting one bash fence so the always-run CONVERGENCE_ARGS construction stays outside the marker. All three flag-absent fallbacks were left outside their markers. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize review and discuss-phase-assumptions Admit state:reviewer-instances-configured (two peripheral notes share it; the core reviewer-lane dispatch stays unmarked — it is the workflow's primary always-evaluated logic, not an optional branch) and state:auto-advance-active, which folds --auto OR two config keys into one boolean so the grammar stays operator-free. discuss-phase-assumptions was the highest-risk edit in this PR. Its auto_advance step is a full if/elif/else; gating it whole would have deleted the flag-absent fallback needed exactly when --auto is off. Split verified exact: resolvers 636-651 and the 'End here' fallback 668-669 both stay outside the marker; only 653-667 is gated. Adds emitted-drift acks for the two files that grew — review.md (+55 B) and autonomous.md (+737 B from 80799211c, which had none and would have red-gated the push. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): fragmentize docs-update, update, transition and new-milestone Part A Completes the 13-workflow rollout. Three of these had no init call at all and gained a dedicated entry point plus their first gsd_run query line. Admits state:is-monorepo and adds state:next-channel, state:workstream-active and state:flat-mode. Vocabulary 26 -> 30 atoms. Part A of new-milestone applies when NO workstream is active — the negation of state:workstream-active. Rather than teach the grammar negation, which is the Greenspun drift the frozen list exists to prevent, it gets a separate positively-phrased atom whose fact is the inverse. Part B, which always runs, stays outside the marker. flag:--verify-only is deliberately NOT admitted: docs-update has no contiguous purely-additive region for it, and an atom without a consuming section is dead vocabulary. Evidence recorded in the slice report. update.md reuses its existing resolved $GSD_TOOLS rather than prepending the canonical preamble, which would have clobbered it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): stop automated-ui-verification re-resolving its own gate, retire dead vocabulary Two defects the new tests caught. The automated-ui-verification step re-ran gsd_run loop render-hooks and recomputed UI_PHASE_ACTIVE inside a body that is only read when that fact is already true — the circular self-disabling pattern this design forbids, introduced by 3c654b168. cmdInitVerifyWork now exposes ui_phase_active and the step consumes it. Its launcher preamble goes too: no gsd_run remains. The Playwright-MCP check stays as prose — that is live session state. Dead vocabulary predating this PR: flag:--full and state:needs-codebase-map were admitted with a gate-1 claim that never materialized. flag:--full is removed, redundant once quick folds it into discuss/research/validate. state:needs-codebase-map gets the real consumer it always lacked, gating new-project's codebase-map offer. Vocabulary 30 -> 29, and no atom is now without a consuming section. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(#2994): add the atom-admission, inversion and resolver-hoist gates The two existing parity guards prove vocabulary/predicate symmetry but never that a fact is computed — an atom no cmdInit* assembles evaluates false forever. These close that hole: - per-atom satisfiability for all 29 atoms, plus an anti-vacuity assertion so the loop cannot silently cover zero atoms - dead-vocabulary check against the shipped manifest - inversion guard: the flag-absent fallbacks in discuss-phase-assumptions and verify-work must stay outside their markers - data-driven resolver-hoist guard over the shipped manifest, so a future extraction cannot reintroduce the circular class - compound-fold coverage (--full, --cross-ai, --rc, config-only --auto) - null-vs-[] degraded/computed distinction, and flag value shapes Also repairs the frozen-vocabulary lock, which was stale and red for the seven atoms earlier commits on this branch shipped. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs(#2994): add changeset for the fragment-model rollout Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * test(#2994): cite the issue on the two new allow-test-rule exemptions ADR-456 requires an issue ref on the same line as the annotation. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * docs(#2994): correct the atom-count claims after retiring flag:--full The vocabulary doc comments still said 30 entries; it is 29 since flag:--full was removed as dead vocabulary. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): dedupe the phase-fallback block and harden --ws parsing Review findings. MAJOR: the three new init entry points each pasted a verbatim copy of the guardedFindPhase/guardedGetRoadmapPhase fallback, taking the repo from four copies to seven — DEFECT.GENERATIVE-FIX. Extracted applyRoadmapFallback and folded six of the seven; each call site keeps its own field-set via a closure. Duplication removed rather than papered over with a parity test. cmdInitPhaseOp stays out: its fallback omits has_reviews, so it is not a byte-identical copy, and it is CRITICAL-radius. LOW, pre-existing: GSD_WS captured [^[:space:]]+ and expands unquoted, so a workstream name holding glob metacharacters would expand against the filesystem. Narrowed to [A-Za-z0-9._-]+. The unquoted expansion is kept — it must word-split into two args and vanish when empty. Also restores the vocabulary ordering convention, and fixes a masked test bug the mandated run surfaced: the flag-forwarding guard checked only the first init line per workflow, but new-milestone has two, so a real failure was reporting exit 0. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): drop the stale new-milestone emitted-drift ack new-milestone.md was acked for a +406 B growth measured against an intermediate commit. Net against origin/next it SHRANK by 8 bytes, so nothing needed the ack and it explained nothing — which the differential attribution check reports as a stale acknowledgment, not a pass. update.md's entry stays: it genuinely grew +703 B. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): resolve the 15 failures from the full matrix run All 15 were real and identical on both lanes. REAL REGRESSION: autonomous.md hit 41479 chars against the #2196 guard's 40960 cap — a CHARS cap distinct from the LARGE tier byte cap, which the five section stubs pushed it over. Extracted the 3a.5 UI Design Contract body to references/; now 39968 chars, and the file nets -795 B vs base, so its growth ack is deleted rather than left stale. REAL DEFECT: docs referenced /gsd-transition, which is not a live registered command. Reworded. STALE FIXTURE: the emission byte-identity test hardcoded two marked workflows; this branch legitimately marks fifteen. Fixture corrected — the source was right. The rest were drift guards over the eight workflows the earlier sweep did not cover, retargeted at where the content now lives with non-vacuity proven by blanking each step file and confirming failure. The GSD_WS forwarding guard was checked as a possible real break and is not one: the charclass narrowing is intact and forwarding works end to end. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): drop the ack for a newly-added reference file A new file's emitted ripple is attributable to the diff that adds it, so the acknowledgment explained nothing and the differential check reports it as stale. Removing the last entry removes the fragment — an empty one signals nothing. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * fix(#2994): retarget the UI-contract guards and clear two transitive advisories The §3a.5 extraction that brought autonomous.md under the #2196 char cap moved its body to references/autonomous-ui-design-contract.md, so ten guards in autonomous-ui-steps and check-ui-safety-gate were asserting it against the host. Retargeted via a combined read, each proven non-vacuous by blanking the reference file and confirming failure. This class had already bitten twice on this branch because each sweep was scoped to the workflows touched at that moment, so this one was exhaustive: ~70 test files across all 13 workflows, zero further broken or vacuous assertions found. Also clears two high transitive advisories the matrix flagged on one lane — fast-uri GHSA-7p8r-x3mc-p8w7 and three ip-address SSRF/trust-boundary issues. Both pre-date this branch: package-lock.json was untouched until now, so the production tree was byte-identical to the base. Lockfile-only, package.json unchanged, verified against a real npm ci install. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> * chore(#2994): backfill changeset pr number to 3030 --------- Co-authored-by: sim <sim@local> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Tom Boucher
|
3f6b063fbb |
chore(#2799): invoke_reviewers and write_reviews iterate declared lanes (#2861)
* chore(#2799): resolve reviewer lanes into executable invocation plans Phase 5b of ADR-2782. Adds the resolver and runner that let invoke_reviewers iterate declared lanes instead of hand-authored per-CLI bash. Five additive descriptor amendments, each forced by a lane that ships today: - LaneHandler gains 'opencode' — the lane rebuilds its review from assistant text parts of a --format json stream; a plain stdout copy re-breaks #1936. - modelConfigKey — antigravity's key is review.models.agy, not .antigravity, so resolving by slug silently dropped a configured model. - defaultHost/fallbackModel — Phase 4 federated every *_host with a default of empty string; the real fallback only existed in the bash. - args becomes an argv template with a closed four-placeholder vocabulary. Positional splicing produced 'codex --model M -o F exec --ephemeral', which is not a valid invocation: codex injects in the middle, twice. - kimi-code lane, with the bounded command-capability probe (needle --output-format) that tells Kimi Code from the legacy python kimi-cli. Parity gate re-pointed: the workflow-text families it scanned are the text this phase deletes, so they are replaced by descriptor-to-registry parity plus an anti-parity check that no bespoke leg returns. jq, curl and external timeout/gtimeout all drop out of the review path. Refs #2782 * chore(#2799): add review-lane query surface and widen the manifest vocabulary Adds the gsd-tools 'review-lane' route (plan/invoke/sections) the workflow loops over, projects all twelve lanes into their capability manifests, and widens capability-validator for the amendments. opencode admitted to VALID_LANE_HANDLERS under the second arm of the enum's own admission rule: one lane, justified by a documented upstream defect data cannot express (#1936 — the agent can end its turn with zero output tokens and --format default then drops the assistant text entirely). Two bugs caught by an end-to-end stub run and fixed here: - loadConfigResolved returns a provenance wrapper, not the config; using it directly resolved every key to undefined, which reads as 'nothing configured' and silently dropped every model override. - hasBinary used shell:true with an args array (Node 26 DEP0190). Replaced with a PATH scan that spawns nothing at all. Refs #2782 * chore(#2799): iterate declared lanes in invoke_reviewers and write_reviews Replaces the eleven hand-authored per-CLI bash legs with a loop over resolved lanes, and renders REVIEWS.md sections from each lane's declared reviewsSection instead of thirteen hardcoded headings. review.md drops from 1104 lines to 507 (61KB to 28.7KB). Parity gate re-pointed, as agreed: the leg-marker and section-heading families scanned exactly the text this phase deletes, so they are replaced by descriptor-to-registry parity in both directions, plus an anti-parity check that fires if a bespoke leg is ever re-added. Enum, emitting sites and the Object.keys lock moved together. The budget-trim helper is hoisted out of the Ollama leg: it was always lane-agnostic, and any lane may now declare a promptBudgetKey. Refs #2782 * feat(#2799): bind the consented egress host and re-verify it at invocation Completes ADR-2782 D5. Rule 1 was recorded in the ADR as delivered by Phase 3 but was not implemented: ConsentRecord had no host field and nothing in the tree bound one, so this phase's rule-4 comparison had no baseline. ConsentRecord gains an OPTIONAL reviewerHost. Optional is the whole design: isValidConsentRecord does not require it, so every record already on disk stays valid and no re-consent storm fires (D4 rule 5). It is deliberately excluded from disclosureSignature — the loader has no config resolver, so folding a config-derived value in would make loader and lifecycle compute different signatures for the same manifest and re-prompt forever. Install resolves hostConfigKey (falling back to the lane's declared defaultHost, which is what the invocation path uses) and records it. Invocation re-resolves and blocks on mismatch rather than silently redirecting. Absence allows: no record, or a record predating the field, means nothing to compare — denying there would break every existing local-model user on upgrade. Refs #2782 * test(#2799): cover the resolver, runner and handlers; retarget the parity suites Adds the golden invocation-plan table (one row per shipped lane, derived from the bash legs rather than the descriptor types) plus runner coverage for the probe, empty-output policy, the three handlers and the egress check. Retargets the existing suites onto the new contract: descriptor-to-registry parity, the anti-parity check, the opencode handler, and the twelfth lane. Two corrections found by running them: - modelConfigKey was required; that breaks D4 rule 2, since a reviewer manifest authored before this phase would fail validation on upgrade. It is optional, read as null when absent. - the antigravity non-zero-exit test pre-seeded the transcript, which asserted that a STALE entry leaks through — the exact bug the watermark prevents. The spawn now appends, as the real tool does. Refs #2782 * fix(#2799): restore agy --add-dir and the self-report prompt in the handler Retargeting the three legacy reviewer suites off the deleted bash surfaced two real regressions in the port, both #2176: - --add-dir was dropped. Without it agy's permission context never receives the cwd repo, so the agent anchors on its own scratch dir and reviews the plan text in isolation — the exact failure the Review Instructions forbid. It is capability-probed, because an older agy rejects the unknown flag outright and a lane that fails to start is worse than one running on the prompt anchor. - the prompt lost the clause mandating a REVIEWED-WITHOUT-REPO-ACCESS self-report, which is what makes a blind review distinguishable from a grounded one. antigravity now builds its own prompt variant. Also ports the #2073 mode-2 cli.log diagnostic, which was dropped: a pinned model that 404s exits 0 with empty stdout AND an empty transcript, so agy's own log is the only evidence that anything failed. The three suites now assert against the plan and the handler instead of matching fence text, so they no longer need allow-test-rule exemptions. Refs #2782 * docs(#2799): document the declared lanes, the new flag, and dropped prerequisites COMMANDS.md gains --kimi-code and replaces the jq-prerequisite paragraph, which is now false: no lane requires jq, curl or an external timeout. Adds the changed-egress-destination behavior, since a blocked lane is something a user can hit. CONFIGURATION.md records that the model config key is declared per lane rather than derived from the flag — antigravity's is review.models.agy — and adds review.models.kimi-code. reviewer-instances.md now routes an instance through its lane's single invocation seam instead of a copied per-adapter bash block, which is what lets a cross-cutting fix reach instances for free. That required implementing the --model/--agent/--as flags it documents; --model re-resolves through the lane's argv template rather than splicing, so the flag lands where the lane declares it rather than ahead of a subcommand. CONTEXT.md glossary gains both new modules. Refs #2782 * chore(#2799): drop the stale emitted-drift acknowledgment The only entry was #2797's, acknowledging COMMENT-ONLY GROWTH in review.md. That file now shrinks by ~32KB and every emitted hash that moved is attributable to this diff, so the ack no longer explains anything. Removing the last entry means removing the file: its presence is the alarm, and an empty one signals nothing. Verified by deleting it and re-running the attribution and provenance gates plus lint:ci — all green without it. Refs #2782 * docs(#2799): record the Phase 5b vocabulary widenings in ADR-2782 Five additive amendments, each forced by a lane that ships today, plus two corrections the phase had to make rather than work around: - D5 rule 1 was recorded as delivered by Phase 3 and was not implemented, so this phase's rule-4 comparison had no baseline. Recorded because an ADR asserting a rule was delivered is exactly what stops a later phase checking. - The DEFECT.GENERATIVE-FIX gate is re-pointed: its workflow-text families scanned the text this phase deletes. Also records that D7's 'skip the probe where no bounding mechanism exists' carve-out is obsolete — in practice it meant the Antigravity lane ran unbounded on every stock macOS host, which ships neither timeout nor gtimeout. Refs #2782 * fix(#2799): close four defects found by adversarial review Two confirmed bugs, both reproduced before fixing: - resolveLanePlan was not total. An openai-http lane with a missing or non-object invoke dereferenced inv.hostConfigKey and threw, contradicting the module's own documented contract; the spawn branch guarded correctly and the http branch did not. The CLI seam resolves every selected lane in one map, so one malformed overlay manifest would have aborted the whole review rather than dropping its own lane. Guarded, plus a per-lane try/catch at the seam so a throw can never take down siblings. - A reviewer-instance model was silently dropped for any lane declaring modelConfigKey null (cursor, qwen, coderabbit). reviewer_instances validates that cli is a known slug but never that the slug accepts a model, so a user could configure one, get a clean run, and never learn a different model reviewed their plan. Now warns explicitly. Two hardening fixes: - The slug is concatenated into artifact paths, so LANE_SLUG_RE is enforced in the resolver rather than inherited from a validator that does not run on this path — the module documents itself as the overlay-manifest trust boundary, so it should not depend on someone else having checked. - normalizeHost mangled a scheme-less value: new URL('localhost:11434') parses with an empty hostname, so it became 'localhost://11434' and was compared and requested as if real. An empty hostname now means not-a-URL. Also documents the one gap that cannot be closed here: the antigravity watermark is keyed by workspace, so two concurrent reviews of the same repo share a transcript. agy exposes no per-invocation id to filter on, so the handler now states which half of its never-stale guarantee actually holds. Refs #2782 * test(#2799): retarget the remaining eight review.md-asserting suites The remote runner found 37 failures the local sweep missed (it hit the shell's two-minute cap before reaching these). All eight extract per-CLI bash from review.md that this phase deletes; each protects a real invariant, so each is retargeted onto the plan, the runner or the handler rather than removed. Three real defects surfaced by doing so: - effort args never reached ANY lane. model-resolver.cjs exports no resolveExecution, so effortFor silently returned [] every time. Restored by calling the same bounded resolve-execution query the bash legs used — and NOT with --raw, which prints the resolved effort rather than the picked field, so claude got 'low' instead of '--effort low'. - the timeout guidance lost 'a silent empty output is a timeout kill, not a crash' — the operator note that exists because of the Codex 0xc0000142 misdiagnosis. Restored. - the opencode handler dropped EMPTY assistant text parts. The shipped jq was , and only substitutes for false/null — an empty string is truthy in jq and contributed a blank line. Found by a property test shrinking to ['', '']. The opencode property suite no longer spawns jq at all, which deletes the #2099 hang mechanism it was architected around rather than mitigating it. Refs #2782 * fix(#2799): register the two new generated modules, and untrack them The remote runner caught build output committed to git. Both new modules compile from src/*.cts into gsd-core/bin/lib/*.cjs, and every sibling generated that way is gitignored and eslint-ignored (ADR-457) - including Phase 1's own review-lane-descriptor.cjs. Mine were neither, so repo-invariants' "each bin/lib/*.cjs is linted xor ignored according to migration state" failed. Registered both in .gitignore and eslint.config.mjs alongside the Phase 1 module, and dropped them from the index. Nothing about the shipped behaviour changes; the artifacts are rebuilt by build:lib. This is the new-.cts-module registration ripple, and it is the one part of it I had not completed - the CONTEXT.md glossary and the inventory manifest were already done. Refs #2782 * chore(#2799): backfill changeset pr number to 2861 * chore(#2799): backfill changeset pr number to 2861 --------- Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
0408276791 |
chore(#2797): federate reviewer config keys off the central schema (#2841)
* chore(#2797): federate reviewer config keys off the central schema Phase 4 of epic #2782 (ADR-2782 D9, config half). Runs AFTER 5a per the ADR's swap amendment: a federated config slice lives inside a capabilities/<id>/capability.json, and three of the five key families had no capability directory until 5a created them. Four key families move to the lanes that use them; the central-schema removal and the federated addition land in this one commit because the exclusivity invariant fails the build on a key present in both. review.max_prompt_tokens, review.default_reviewers and review.reviewer_instances describe policy ACROSS lanes and stay central. Two things the issue did not name, both found while building it: 1. THE EXCLUSIVITY GATE WAS BLIND TO PATTERNS. It compared federated keys against manifest.validKeys only, and two of the four families (review.models.<slug>, review.max_prompt_tokens_per_reviewer.<slug>) were pattern-backed. That is not cosmetic: isCentralConfigKey consults those patterns and mergeFederatedConfig skips every key for which it returns true, so declaring a slice while the pattern survived would have shipped an INERT slice behind a green gate — the exact half-migrated shape the invariant exists to prevent. The gate now loads the patterns from the same manifest the runtime reads. 2. AN UNSET PER-LANE BUDGET NOW RESOLVES TO 0, NOT NOT-FOUND, because a federated key always resolves to its declared default. The three fallback guards in review.md checked only empty-or-"null", so a user who set the GLOBAL review.max_prompt_tokens would have silently lost trimming on the HTTP lanes. The guards now treat 0 as unset. D9 says review.models.<slug> is owned by "the lane whose slug it names". That is false for one lane: the shipped key is review.models.agy while the slug is antigravity. Ownership follows the lane; the key name is preserved, because renaming would break every config that sets it. Existing tests updated rather than left asserting the old world: config-get on a cleared federated key yields empty instead of not-found (what #2046 actually protects — never persisting the literal "null" — is unchanged and still asserted); the config-schema dynamic pattern representative moves to reviewer_instances; the prototype-pollution guard case moves to a surviving dynamic prefix so alert #26 keeps its coverage, with a new case asserting the old key is now rejected earlier; and Phase 2's harvest-widening inertness assertion becomes an ownership assertion, since Phase 4 is what consumes it. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): use a -1 sentinel so an explicit per-lane budget of 0 survives A federated config key always resolves to its declared default, so an unset per-lane prompt budget needed a value the workflow could treat as 'not configured'. The first cut used 0 — which is wrong: 0 is already a LEGITIMATE per-lane budget meaning 'do not trim this lane' (the early-return guard in prepare_trimmed_prompt_for_reviewer). Treating it as unset would have silently switched a user who deliberately disabled trimming for one lane onto the global budget. The sentinel is now -1, which is not a valid token budget, so all three states stay distinguishable: unset falls back to global, an explicit 0 disables trimming for that lane, and an explicit N is used. Locked by three CLI round-trip tests. Surfaced by the isolated security reviewer before it crashed mid-run; verified independently against the shipped trim guard rather than taken on trust. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): update central-registration assertions and stay under the review.md cap The remote runner caught both; my local sweep missed the files. 1. tests/plan-review-convergence.test.cjs asserted the three local-server host keys are in VALID_CONFIG_KEYS. They are federated to their lane capabilities now, and the exclusivity invariant forbids a key living in both places. What #2306-local actually protects is that config-set ACCEPTS them, so that is what is asserted — via isValidConfigKey, the predicate config-set itself uses, which spans central and federated. A second assertion pins federated ownership, so a silent reversion back to the central schema fails too. 2. review.md exceeded the LARGE tier hard cap (62583 > 61440). That cap is a red line, not a budget to raise. The three per-lane budget guard comments were near-identical; condensed to one terse line each. 61371 bytes, 69 to spare. Real extraction to workflows/review/modes/ is Phase 5b/6 work. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2797): fail closed on a broken config-schema manifest; reconcile stale docs Isolated security review findings. MAJOR — loadCentralConfigPatterns failed OPEN. It swallowed a JSON parse error and returned [], while its sibling loadCentralConfigKeys, reading the SAME file, writes to stderr and throws ExitError(1) on that identical failure class. Fail-open here defeats the gate this function exists to feed: with zero patterns, validateCrossCapability's pattern-collision check silently passes and an inert federated slice ships green. It was masked in the one production call site only because loadCentralConfigKeys runs first against the same path — a coincidence of ordering, not a guarantee, and this function is exported and called standalone. The two now share a contract: ENOENT is the legitimate absent case, anything else throws loudly. A single unparseable PATTERN is still skipped, which degrades to "checked less" rather than blocking every build. The branch had zero coverage; it now has two tests (malformed JSON, EISDIR). MINOR — docs/CONFIGURATION.md still listed review.models.qwen and review.models.cursor as settable, ~770 lines below this PR's own new Ownership section. Those lanes take no model flag, so they declare no model key and config-set now rejects them. Rows removed; the missing review.models.agy row added; the per-reviewer budget row corrected to name only the lanes that own a budget key, and to document that a per-lane 0 disables trimming for that lane. Also fixes a shadowed "raw" binding introduced by the fail-closed change, which made the generator unrequirable — caught immediately by its own --check. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2797): backfill changeset pr number to 2841 * fix(#2452): make the base-ref mutation test hermetic against leaked GIT_* env tests/mutation-workflow-base-ref.test.cjs fails on PR branches while next stays green, and it is currently blocking at least three unrelated PRs (#2841, #2832, #2827) with: error: invalid object 100644 <sha> for 'base-N.txt' error: Error building trees The existing loop comment attributes this to `git add .` rehashing O(n^2) blobs "before the object write had landed" and works around it by staging one path per iteration. That is not the cause: sequential execFileSync calls cannot race each other's object writes, and the failure persisted after that change — it simply moved to a lower commit index. The cause is that the git() helper inherited the runner's environment. A leaked GIT_INDEX_FILE makes `git add` write into a DIFFERENT repository's index; GIT_OBJECT_DIRECTORY / GIT_ALTERNATE_OBJECT_DIRECTORIES send the blob to another object store; GIT_DIR / GIT_WORK_TREE redirect the whole operation. In every case `git commit` then cannot resolve a blob it just staged, which is precisely the error above. Verified by negative control: with GIT_DIR exported, this test fails on the unfixed helper (the git commands operate on the wrong repository entirely); with the helper stripping GIT_* it passes. The single-path staging is kept — it is genuinely less work — but it is no longer load bearing. Found while shipping #2797. Fixed in place rather than deferred: it is a defect surfaced during the work, and it is blocking other contributors. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2452): build the base-advance commits empty, removing the lost-object class The base-ref guard has been failing in CI with: error: invalid object 100644 <sha> for 'base-N.txt' error: Error building trees It is currently red on at least three unrelated PRs (#2841, #2832, #2827) while next stays green. Two theories have now been tried and neither held. #1881 blamed `git add .` rehashing O(n^2) blobs and switched to staging one path per iteration; the failure moved from commit 32 to commit 25 and carried on. The preceding commit here made the git helper hermetic against leaked GIT_* environment — that IS a real vulnerability (with GIT_DIR exported the helper operates on the wrong repository entirely, proven by negative control) but it produces a different error than CI reports, so it is not demonstrably the cause either. Neither trigger reproduces off-CI, so this stops guessing at the trigger and removes the failure CLASS instead. The loop needs base-branch DEPTH and nothing else: no assertion reads these commits' contents, and base-side files cannot appear in `origin/base...HEAD` regardless. `--allow-empty` writes no blob and no tree, so there is no object for the index to reference and lose. It is also far less work than 60 write+hash+index cycles. The guard still proves its mechanism: the test asserts that a --depth=1 base fetch FAILS and a full fetch resolves, so a broken topology would surface immediately rather than passing vacuously. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> Co-authored-by: Test <test@example.com> |
||
|
Tom Boucher
|
8b44a0da43 |
chore(#2794): single-source the reviewer invocation contract + parity assertion (#2820)
* chore(#2794): single-source the reviewer invocation contract Phase 1 of epic #2782 (ADR-2782). Introduces one core descriptor table as the declared contract for all 11 cross-AI reviewer lanes, and the DEFECT.GENERATIVE-FIX parity assertion the roster has never had. The lane contract lived in three unrelated surfaces — the roster, ~640 lines of hand-authored per-CLI bash in invoke_reviewers, and the write_reviews section headings — so cross-cutting fixes landed per-leg (#2494 and #2605 were the same empty-output defect filed twice). - src/review-lane-descriptor.cts: frozen table declaring per lane the slug, flags, probe, invoke shape, timeout floor, empty-output policy, REVIEWS.md section, evidence class, required binaries, prompt-budget key and handler. Field names track ADR-2782 D1/D2/D6/D7 verbatim so Phase 2 harvests the shape with no translation layer. It declares; it does not execute — invoke_reviewers iterates in Phase 5b. - checkReviewerLaneParity: bidirectional parity across descriptor, roster, invoke_reviewers legs and write_reviews sections. Forward-only would miss the failure it exists to catch (#2718 added a leg, #2781 was the drift). ADR-1517 instance headings are exempt per D8. - Legs carry an explicit <!-- reviewer-lane: slug --> marker; five non-lane bold labels share the bold-then-fence shape a heuristic matcher would key on. - ADR-2782 D4: an explicitly-flagged reviewer that cannot run is now an error in both the core module and the workflow prose that mirrors it. A code-only change would be unobservable — the module has no production caller; the workflow narrates the policy. Discovery paths (--all, review.default_reviewers) stay lenient. - Fixes the qwen leg, the last one discarding stderr to /dev/null. Two ADR-2782 D2 vocabulary widenings were forced by surveying the shipped legs: promptChannel 'none' (CodeRabbit is fed no prompt) and outputChannel 'file-arg' (Codex writes via -o and discards stdout, #1698). Both are additive and closed; Phase 2 owns the validator. Closes #2690 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2794): make the parity checker total and pin the lane slug grammar Findings from the orthogonal review passes. Spec axis — the module claimed its vocabulary tracked ADR-2782 D1/D2 "verbatim" while diverging in three undisclosed ways, which is the translation layer Phase 2 was supposed to be spared: - `transport` moves from `invoke.transport` to the LANE level, a sibling of `probe`/`invoke`, exactly as D1's manifest example places it. The nested form read better as a TS discriminated union; the union is now discriminated at the lane level instead, which costs nothing. - The header and the CONTEXT.md glossary now enumerate all FOUR widenings (adding `outputArg` and `flags[]`), not two. Standards axis — CLAUDE.md requires a fast-check property test for a parser, and `checkReviewerLaneParity` parses markdown for markers and headings. Adding one found two real defects that the hand-written matrix missed: - NOT TOTAL: a malformed descriptor entry threw on `lane.flags` iteration, contradicting the module's own "never throws" claim. Every field is now narrowed from `unknown` at the trust boundary and reported as MALFORMED_LANE / INVALID_SLUG. This matters because Phase 2 feeds this function third-party overlay data, and a parity gate that crashes is indistinguishable from one never run. - SILENT GRAMMAR MISMATCH: LEG_MARKER_RE captures only [a-z0-9_-], so a slug outside that class was unmatchable — its marker could be present and correct and the scan would still report LEG_MARKER_MISSING forever. LANE_SLUG_RE now pins the grammar and a violating slug is reported INVALID_SLUG. A loud named violation beats a silent miss. Generators are document-shaped, not writer-seeded (CONTRIBUTING #2371): seeding from the module's own matchers could only produce documents those matchers already recognize. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * fix(#2794): register the new bin/lib module in the ESLint ignore list The remote runner caught this; lint:ci did not, because the invariant lives in the test suite rather than the lint chain: tests/repo-invariants.test.cjs "each bin/lib/*.cjs is linted xor ignored according to migration state" -> tsc-generated bin/lib modules not yet added to ESLint ignore list: review-lane-descriptor.cjs Adding a src/*.cts module ripples to six surfaces (.gitignore, the ESLint ignore list, docs/INVENTORY-MANIFEST.json, the CONTEXT.md glossary, the capability/inventory manifests, and any size baseline). The other five were covered; this was the miss. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * docs(#2794): amend ADR-2782 D1/D2/D8 with the vocabulary Phase 1 surfaced Building the Phase 1 descriptor table against all eleven shipped legs is the first time every lane's contract was written in one place, and it surfaced four cases the ADR's original survey did not cover. Amending the design lock rather than diverging from it, so Phase 2 (#2795) implements the manifest validator against the amended vocabulary instead of rediscovering the gaps. All four are additive widenings of closed enums; no decision reverses: - D2 promptChannel gains `none` — coderabbit is fed no prompt at all, it reviews the working-tree diff. - D2 outputChannel gains `file-arg` — the ADR called a file-writing lane a shape a real CLI *could* take; codex already is one, writing via -o/--output-last-message and discarding stdout (#1698). - D2 gains `outputArg`, required iff file-arg — knowing the review lands in a file is useless without the argument naming it. - D1 `flag` becomes `flags[]` and D8's uniqueness flattens across lanes — antigravity is selected by both --antigravity and --agy, which a single-valued field cannot express. This is the same evidence path that produced the openai-http transport: the vocabulary widens on a lane that exists, under review, never on speculation. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * chore(#2794): backfill changeset pr number to 2820 --------- Co-authored-by: Claude Opus 5 <noreply@anthropic.com> |
||
|
Tom Boucher
|
2c44241a0b |
fix(#2605): make dropped local-server reviewer lanes loud instead of silent (#2689)
* fix(#2605): make dropped lm_studio / llama_cpp reviewer lanes loud, not silent The `lm_studio` and `llama_cpp` reviewer legs in `gsd-core/workflows/review.md` carried the same empty-output defect as the claude/gemini legs (#2494, fixed in #2592) in a worse variant: when the local endpoint was unreachable or returned empty content, they wrote NOTHING to `{run_dir}/gsd-review-<leg>.md`. There was no `[ ! -s … ]` stub at all, so the file never existed, `write_reviews` omitted that reviewer's section, and the outcome was indistinguishable from the reviewer never having been selected. Two diagnostic holes are closed, because an OpenAI-compatible server fails in two ways that leave evidence in different places: - Transport failure (endpoint unreachable): curl writes to stderr and exits non-zero. Both legs used `curl -s`, which suppresses curl's ERROR text as well as the progress meter, and then discarded stderr to `/dev/null` — so there was nothing to capture even in principle. Now `-sS` with stderr to a `.err` sidecar, matching the claude/gemini/codex legs. - Application failure (HTTP 4xx/5xx): curl exits 0 and the error JSON is in the response BODY, so stderr is empty and only the body is evidence. The stub appends the raw response. The `llama_cpp` leg additionally piped curl straight into `jq`, throwing the body away before anything could inspect it; the response is now captured to a variable first, as the `lm_studio` leg already did. The existing `>&2` warning is kept and now points at the stub file. Failing-first verified by extracting both shipped blocks and running them under a real bash against a stubbed curl: pre-fix, all three failure modes produce NO review file and no `.err` sidecar for both legs; post-fix, each produces a diagnosable stub, and a successful review still passes through untouched. Closes #2605 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QT3ibz5qJuDuGqpTGRYVGf * fix(#2605): close the remaining silent-drop paths the review surfaced Four defects found by the orthogonal review of the first commit, all fixed here rather than deferred (they are the same defect class the issue is about, and three of them sit inside the code that commit touched). 1. CodeRabbit leg was still unguarded. `coderabbit review --prompt-only 2>/dev/null > file` with no `[ ! -s … ]` stub — the last leg still shaped like pre-#2494 code. A missing or unauthenticated binary left a zero-byte file that write_reviews rendered as "ran cleanly, nothing to report". It now captures stderr to a `.err` sidecar and emits the same diagnosable stub as every other leg. This is the leg the next issue in the #2494 -> #2592 -> #2605 series would have been about. 2. The budget-skip path dropped the lane just as silently. When `prepare_trimmed_prompt_for_reviewer` fails, `*_SKIP=1` bypasses the entire block — guard included — so no file was written and the only trace was a stderr warning nothing persists. All three local-server legs now write a "review skipped: prompt budget too small" stub on that path. 3. Whitespace-only replies evaded the guard. `[ ! -s … ]` counts BYTES, and command substitution strips trailing newlines but not spaces, so a reply of `" "` was written out and passed as a "successful" but vacuous review — the same indistinguishable-from-success outcome the guard exists to prevent. A `case` glob now normalizes whitespace-only content to empty. 4. `echo "$VAR"` swallowed option-like content. bash's builtin `echo` treats a value of exactly `-n`/`-e`/`-E` as a flag and writes 0 bytes, which would trip the empty guard and DISCARD a genuine reply. Switched to `printf '%s\n'`, the idiom the OpenCode leg in this same file already uses for this reason. Also brings the Ollama leg to parity while it is in hand: it always emitted a stub so it never silently vanished, but it was the least diagnosable of the three local-server legs — bare `-s`, stderr to /dev/null, and the response piped straight into jq so the error body was discarded unread. Verified by extracting all four shipped blocks and running them under a real bash against stubbed CLIs: 22 cases (7 per local-server leg x 3, plus CodeRabbit) all produce the contracted output, and a successful review still passes through untouched on every leg. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QT3ibz5qJuDuGqpTGRYVGf * test(#2605): regenerate golden install-parity fixtures for the review.md edit The remote test run on the prior commit failed with 19 "golden parity — <runtime>" mismatches. review.md is installed into every runtime's tree, so editing it changes its content hash in all 19 golden fixtures. Regenerated with `npm run gen:golden`; the diff is exactly one line per fixture — the gsd-core/workflows/review.md hash — and nothing else. This is the second ripple of a workflow edit, alongside tests/workflow-size-baseline.json which the first commit already updated. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01QT3ibz5qJuDuGqpTGRYVGf * chore(#2605): backfill changeset PR number (#2689) --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Tom Boucher
|
920a5f3f06 |
fix(#2589): use --raw/--pick for config/model/verify lookups, drop jq dep (#2673)
* fix(#2589): use --raw/--pick for config/model/verify lookups, drop jq dep The reviewer/workflow config lookups resolved scalars and object fields with a `gsd_run query <cmd> … | jq … 2>/dev/null || <default>` shape. On any machine without jq (the default on Windows/Git-Bash) the jq stage fails with exit 127, the failure is swallowed by 2>/dev/null + the trailing || default, and the variable comes back EMPTY — the configured per-lane model/host/budget is silently dropped and the lane falls back to CLI defaults with no diagnostic. gsd-tools ships native flags that do the same job with no external dep: config-get <key> --raw (strips JSON quotes off a scalar) resolve-model <id> --pick model (descends an object) resolve-execution … --pick <f> (same) verification.status … --pick status Replaced every jq-piped config/model/verify lookup across review.md (×23), plan-phase.md, ship.md, debug.md (incl. the redundant boolean coercion — --raw returns true/false as bare tokens natively), autonomous.md (×2), ai-integration-phase.md (×4), and eval-review.md. The legitimate structured-JSON jq sites that parse HTTP curl responses (.choices[0], jq -rs, jq -n --rawfile) are untouched — only the jq-replaceable lookups moved to the native flags. Adds tests/fix-2589-config-get-no-jq.test.cjs: a source-invariant guard asserting no audited workflow pipes config-get/resolve-model/resolve-execution/verification.status to jq (fails-first on the pre-fix text, passes after). * test(#2589): update autonomous-converge jq assertion to --pick; regen golden fixtures Two test consequences of the workflow-doc edits in the prior commit: 1. tests/autonomous-converge.test.cjs pinned the OLD jq-dependent shape (`verification.status … | jq -r '.status//empty'`) as the canonical routing contract. The test's INTENT is correct (route human validation through canonical verification.status) but it over-specified the MECHANISM (the jq pipe). Updated the assertion to match the new native --pick status shape; the contract being guarded (canonical verification.status read before the human_needed branch) is unchanged. 2. The golden-install-parity fixtures (19 runtimes) record a content hash of every installed workflow .md; the 7 edited workflows changed those hashes. Regenerated via `npm run gen:golden` (the test's own failure message instructs this). Only the 7 edited workflow hashes changed in each fixture. * fix(#2589): declare jq a prerequisite for the lanes that still need it; repair test file Three defects in the first cut of the #2589 fix: 1. tests/autonomous-converge.test.cjs was a JavaScript syntax error. The regex literal /...2>\/dev/null .../ left the second slash unescaped, terminating the literal early and parsing `null` as regex flags: SyntaxError: Invalid regular expression flags The whole file failed to load, so every assertion in it — including the #1522 and #1526 guards — silently stopped running. Replaced with the string-compare form already used at line 202 for the sibling shell-snippet assertion. 2. lint:ci failed. tests/fix-2589-config-get-no-jq.test.cjs buckets into the capped `config` production module via its `config-get-...` effective prefix, making it a novel offender against the 2-file cap. The test is about workflow documents, not the config module, so it is renamed to fix-2589-workflow-jq-dependency.test.cjs (free prefix) rather than growing the allowlist with a module that does not actually need a 5th test file. 3. The fix deleted the repo's only jq-prerequisite declaration. review.md:244 ("install jq if missing") was the anchor plan-review-convergence.md cites by line number, and it went away with the jq pipes — while the ollama, lm_studio, llama_cpp, opencode, and agy lanes still hard-require jq to parse HTTP /v1/chat/completions responses, opencode's JSONL event stream, and agy's conversation cache. On a jq-less host those five lanes swallow exit 127 into empty output: the same silent-degradation class #2589 exists to close. detect_clis now probes jq alongside the other prerequisites and emits jq:available / jq:missing, and the five dependent lanes are treated as undetected when it is absent, with an install hint. The six lanes that do not need jq stay selectable. plan-review-convergence.md now cites the section by name instead of a line number that moves. Regression guards added to the renamed test file: review.md must keep the jq probe and must name all five dependent lanes, and no workflow may cite review.md by line number. Workflow-size baseline and the 19 install-parity goldens regenerated for the review.md / plan-review-convergence.md edits. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015TCwhbMuY37DzRMCfzTABJ * fix(#2589): decide the ship verification gate on a single verification.status read Isolated review finding (medium). Pre-fix, ship.md captured verification.status ONCE into $VERIFICATION and picked status / next_action / next_command off that cached JSON with three jq calls. --pick takes a single dot-path field, so the mechanical conversion issued three separate queries up front: three node spawns that each re-read the phase VERIFICATION.md and re-derive the commit-time vs mtime staleness comparison, on every ship — including the common passing path that never uses the two message fields. It also meant the gate's verdict and the message shown to the user were derived from three reads with no guarantee they observed the same state. The gate now reads `status` once and decides. The two message-only fields are read on the blocking path only, after PHASE_VERIFICATION_INCOMPLETE is already determined — so the passing path costs one query instead of three, and a concurrent write between reads can no longer make the gate and its message disagree, because the block/allow decision no longer depends on them. Adding a multi-field --pick to gsd-tools would have collapsed this to one query, but that changes the flag's output contract and belongs in its own change. Regression guard in tests/fix-2589-workflow-jq-dependency.test.cjs: ship.md must read verification.status exactly three times total, the block decision must follow the status read, and next_action / next_command must both appear after the blocking prose so they cannot drift back onto the passing path. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015TCwhbMuY37DzRMCfzTABJ * docs(#2589): document the jq prerequisite for the five reviewer lanes that need it /gsd-review's ollama, lm_studio, llama_cpp, opencode, and agy lanes parse JSON GSD does not produce (OpenAI-compatible /v1/chat/completions responses, OpenCode's JSONL event stream, Antigravity's conversation cache), so they require jq on PATH. Nothing in docs/ said so. Records which five lanes need it, which six do not, that reading configured models/hosts/budgets no longer requires jq at all, and what /gsd-review now does when jq is absent. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_015TCwhbMuY37DzRMCfzTABJ * chore(#2589): backfill changeset pr number (#2673) --------- Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
Behruz Nassre Esfahani
|
a40ee8a5e7 |
fix(#2479): drop the codex hook-trust bypass flag and its capability probe (#2536)
* fix(#2479): drop the codex hook-trust bypass flag and its capability probe The /gsd-review codex lane emitted a hook-trust bypass flag via a capability-probed variable (#1115). Host-harness safety classifiers deny commands carrying the flag (23/33 sampled invocations), one denial citing the probe itself as intent, while flagless retries succeeded 32/32 — the flag only bypasses persisted hook trust, a first-run condition with no steady-state value. Remove both the flag and the probe per maintainer direction (no config key). #1115's diagnosability half — stderr to .err, folded into the lane on empty output — is untouched; its version-gate becomes vacuous with no flag to gate. The regression test inverts: the literal flag is now banned file-wide in review.md (covers continuation lines, carrier variables, and probes), alongside bans on the carrier variable and any codex help-grep probe shape. Fixes #2479 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(#2479): add changeset for PR #2536 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * chore(#2479): changeset body in house format (bold lead) + post-rebase fixture regen Review round 2 Minor: wrap the changeset lead clause in the required **bold** span (reviewer-supplied text, applied verbatim). Rebased onto current next; golden-install-parity (19 runtimes) + size baseline regenerated — delta confined to review.md's hash/size. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Claude Fable 5 <noreply@anthropic.com> |
||
|
0xdhx
|
bd18a593a9 |
fix(#2494): capture stderr and guard empty output on the claude and gemini reviewer legs (#2592)
* fix(#2494): capture stderr and guard empty output on the claude and gemini reviewer legs The gemini and claude blocks in gsd-core/workflows/review.md were the only two of the ten prompt-fed reviewer legs with both `2>/dev/null` and no empty-output guard. Any failure that wrote no stdout — CLI missing, unauthenticated, rate-limited, crashed — left a zero-byte review file with the only diagnostic evidence already discarded. write_reviews substitutes each file's raw content into a `## <Reviewer> Review` section with no marker separating "empty/failed" from "ran cleanly, nothing to report", so a failed lane silently degraded the advertised N-reviewer consensus to N-1 while present_results reported success. Both /gsd:review and /gsd:plan-review-convergence share the invoke_reviewers step, so both were affected. Both legs now redirect stderr to a `.err` sidecar and write a diagnostic stub with the captured stderr appended when the review file comes back empty — the same shape the codex and cursor legs already use. Scope limit, noted in the block comment: the guard only runs if the block itself completes. A host Bash-tool timeout that kills the whole block skips it, the same hard bound the OpenCode block already documents. The existing timeout guidance (#2194) covers that case and is unchanged. Regression test extracts the two dispatch blocks verbatim from the workflow and runs them under bash against a failing CLI stub, asserting the review file is non-empty and carries a diagnosable message plus the captured stderr. It fails against pre-fix review.md (4 of 5 cases). Golden install-parity fixtures and the workflow size baseline are regenerated: one review.md hash per fixture, one size entry. Fixes #2494 * fix(#2494): stamp the changeset fragment with the filed PR number The fragment was authored with the documented `pr: 0` placeholder because the PR number does not exist until `gh pr create` returns. Now that this PR is #2592, stamp it — scripts/changeset/parse.cjs requires pr > 0, so the placeholder would fail the Changeset Required check. |
||
|
Tom Boucher
|
09b535ac00 |
feat(#2481): add a negotiated effortSurface axis and wire invocation-time effort
ADR-1239 gains a ninth negotiated axis, effortSurface (argv | none), declaring how
a host accepts reasoning effort. ADR-443 is amended in the same change because its
recorded deferral is what the axis resolves: its Unblock condition offered paths
(a) and (b) and stated the choice was 'a maintainer call this file records but does
not make'. Path (a) is selected and satisfied here.
Before this, effort reached a runtime only through install-time channels
(EFFORT_RENDERING's frontmatter/api), so reviewer CLIs spawned as subprocesses
silently inherited whatever effort sat in the user's own global CLI config. The
review lane now resolves one universal effort through the ADR-443 cascade and
renders it per host through the negotiated descriptor.
Every per-host value is documentation-sourced, never inferred:
- claude argv -- verified via 'claude --help' (--effort <level>)
- opencode argv -- verified via 'opencode run --help' (--variant)
- codex argv -- codex-rs/exec/src/cli.rs: model_reasoning_effort is NOT a CLI
flag (config.toml key only), so the global -c override is the
only argv route
- 15 hosts undocumented -- their docs state no reasoning setting; the sentinel
fails closed rather than inheriting a profile baseline
No config-file vocabulary member: the only host that ever had one (Gemini CLI's
thinkingConfig) was removed as a sunset runtime by
|
||
|
Tom Boucher
|
40ce95f882 |
fix(#2358): scope review.md and ship.md temp files to a per-run mktemp directory (#2433)
* fix(#2358): scope review workflow temp files to a per-run mktemp dir /gsd-review wrote every prompt/section/output temp file to a hardcoded /tmp path keyed only on the phase number, so two GSD projects sharing a small phase number collide on the exact same path and a crashed run's leftover file becomes bait a later, unrelated run can silently read. ship.md's external peer-review stderr capture was strictly worse — one shared, unqualified path across every project/phase/run. Thread a single mktemp -d "${TMPDIR:-/tmp}/gsd-review.XXXXXX" run directory through every review.md temp path (67 sites) via a new {run_dir}/$RUN_DIR placeholder, mirroring the existing {phase} substitution mechanism, and clean it up at the end of the run. Route ship.md's stderr capture through a per-run mktemp file the same way. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * chore(#2358): regenerate fixtures + lint gate-prep * fix(#2358): repair failing tests after gate verification * fix(#2358): thread RUN_DIR scoping into reviewer-instances.md (#1517) review.md's own invoke_reviewers step lazily loads gsd-core/references/reviewer-instances.md for the review.reviewer_instances codepath, but that doc was missed when review.md and ship.md were moved to the run-scoped {run_dir} temp directory. It still read the combined prompt from the old /tmp/gsd-review-prompt-{phase}.md (which build_prompt no longer writes, breaking reviewer-instances functionality outright) and wrote each instance's output to the old unscoped /tmp/gsd-review-${INSTANCE_NAME}-{phase}.md, leaving the exact cross-project temp-file collision bug open for that code path. Both paths now thread through {run_dir}, matching every other reviewer block in review.md. Extends the existing #2358 regression test with assertions pinning reviewer-instances.md's prompt read and output write to {run_dir}, and adds the Fixed changeset fragment. Regenerated the golden-install-parity content hashes for reviewer-instances.md via `npm run gen:golden` (paths unchanged; only the modified file's hash moved). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * chore(#2358): backfill changeset pr (#2433) --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Cody Anderson
|
98e4233ce9 |
fix(#2176): ground the Antigravity reviewer in the repo under review (#2184)
* fix(#2176): ground the Antigravity reviewer in the repo under review - capability-probe --add-dir (mirrors the Codex bypass-flag probe) and pass the repo root on both invocation arms - anchor _AGY_PROMPT to the absolute repo root; mandate a REVIEWED-WITHOUT-REPO-ACCESS self-report when the repo is unreadable - stamp a [reviewed-without-repo-access] marker on self-reported or scratch-anchored output; Consensus Summary down-weights marked reviews - apply the same absolute-root anchor to the cursor-agent prompt (AC5) Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * docs(#2176): changeset fragment for PR #2184 Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * fix(#2176): review fixes — size baseline, cursor root anchor, anchored blind tells - regenerate tests/workflow-size-baseline.json for review.md's growth - cursor anchor uses git rev-parse --show-toplevel (bare pwd resolved the wrong root from a repo subdirectory) - blind-review tells anchored: self-report to the first lines of output, scratch tell to a workspace-declaration phrasing — a grounded review quoting either string is no longer mis-stamped Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * fix(#2176): round-2 review fixes — scratch-tell bridge, behavioral test, changeset Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * test: regenerate golden-install-parity fixtures for the review.md change Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * test(#2176): pass the transcript path to bash with forward slashes The behavioral detection test substitutes a mkdtemp path into the bash compound; on Windows runners that path contains backslashes, which bash strips, so the transcript is never found and the first assertion fails (windows-latest/24 lane). Git Bash accepts D:/-style paths. Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * fix(#2176): use /gsd:review namespace syntax in workflow comment The slash-command namespace invariant (#3443) bans retired /gsd-<cmd> references in Claude-facing sources; a cursor-anchor comment used /gsd-review. Size baseline + golden fixtures regenerated for the byte change. Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * test(#2176): derive the POSIX path via path.sep, not a hardcoded separator Review finding: out.replaceAll('\\', '/') hardcodes both separators; use the separator-safe out.split(path.sep).join(path.posix.sep) idiom. Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg * test(#2176): use the merged toPosixPath seam for the bash path Per maintainer note: #2247's shell-command-projection now centralizes running-OS → POSIX path conversion; import it instead of the inline split/join idiom. Claude-Session: https://claude.ai/code/session_01Hme55Pvq6BhpgwBcyC5HAg |
||
|
Tom Boucher
|
24c324cbfa |
fix(#2194): add Bash timeout guidance for prompt-fed reviewers in review.md
The Gemini, Claude, and Codex reviewer blocks invoked the CLIs with no explicit timeout, so each inherited the host default (~2 min on Claude Code). A source- grounded review of a large plan set takes ~570s (Codex xhigh) / ~525s (headless Claude) — both exceed that window, so the lane is killed mid-review, its output is empty, and the cross-AI review silently proceeds with fewer lanes. CodeRabbit and OpenCode already documented a timeout; the four main lanes did not. Add a shared timeout-guidance note directing a high Bash timeout (>= 900000; 1200000 for Codex xhigh / headless Claude), referencing BASH_MAX_TIMEOUT_MS for the Claude Code host cap, and framing a slow-lane empty output as a timeout kill (not the 0xc0000142 crash it gets misdiagnosed as) so operators re-run with more time instead of diagnosing a CLI failure. Closes #2194 Recaptures the 18 golden-install-parity fixtures + workflow-size baseline (only the review.md entry changed in each; review is LARGE-tier, 47168 < 61440). |
||
|
Tom Boucher
|
dbc730d8de |
fix(#2073): capability-probe external killer (timeout/gtimeout) for macOS
Code review (HIGH): a hardcoded 'timeout 600 agy' fails with rc 127 on stock macOS (no GNU timeout/gtimeout), silently losing the agy reviewer. Probe for 'timeout'/'gtimeout' via command -v and fall back to agy's native --print-timeout alone when neither exists (mirrors scripts/base64-scan.sh). External cap (600s) stays >= --print-timeout (540s) so it only backstops a pre-session stall. Factor the prompt into _AGY_PROMPT to avoid duplicating the long -p string across both branches. Update the agy + #687 tests to assert the probe + bound + fallback, regen the 17 goldens + size baseline, refresh the maintainer-note version stamp to 1.0.16. |
||
|
Tom Boucher
|
af9069f865 |
fix(#2073): harden agy reviewer block (arg overflow, 404, pre-session stall)
Three failure modes on agy 1.0.16, all fixed by mirroring the Cursor block's
invocation discipline:
* file-reference prompt instead of inline "$(cat)" — a large review prompt
overflowed the exec arg list (rc 126).
* external 'timeout 600' wrapper — --print-timeout cannot fire before agy
creates a session, so a pre-session stall hung unbounded.
* --model from review.models.agy when set — escape hatch for a pinned model
that 404s (exit 0, empty stdout + transcript).
* stdin </dev/null so agy never blocks on a tty.
Also enrich the Step 3 empty-output stub to grep agy cli.log for a
model-availability diagnostic, and correct the stale 'no --model flag' note
plus the 'review.models.agy reserved for future' comment (the config key was
already read but never passed through).
|
||
|
Tom Boucher
|
a62079b2da |
fix(#1865): runtime launcher honors CLAUDE_CONFIG_DIR (#2024)
* fix(#1865): runtime launcher honors CLAUDE_CONFIG_DIR The gsd_run preamble resolved the Claude global install only at $HOME/.claude/gsd-core/bin/, but the installer honors CLAUDE_CONFIG_DIR — so a global install redirected via CLAUDE_CONFIG_DIR was invisible to every gsd_run call (every command failed with 'gsd-tools.cjs not found'). The Claude resolver arm now uses ${CLAUDE_CONFIG_DIR:-$HOME/.claude}, matching the installer + the other runtimes' ${VAR:-default} pattern. Default $HOME/.claude behavior is unchanged. - _runtime-launcher.snippet.sh: Claude arm honors CLAUDE_CONFIG_DIR. - sync-runtime-launcher.cjs re-run: 95 workflows/agents re-synced. - review.md / discuss-phase.md: trimmed to stay under their byte budgets. - runtime-launcher-parity.test.cjs: (A) substring updated for the new form + explicit #1865 assertion that the snippet honors CLAUDE_CONFIG_DIR. - goldens + size baselines recaptured. Closes #1865 * docs(#1865): backfill changeset pr 2024 |
||
|
Behruz Nassre Esfahani
|
23254ca5a7 |
fix(#1936): reconstruct OpenCode review from JSON events (#1992)
* fix(#1936): reconstruct OpenCode review from JSON events; diagnosable empty-output stub On a large review prompt, OpenCode's default `build` agent runs a few read tool calls then ends its turn with zero output tokens (reason:"stop", output:0), so `opencode run --format default` emits empty stdout. The reviewer block redirected stderr to /dev/null and wrote a generic "failed or returned empty output" stub — so the phase silently lost its second independent reviewer with no diagnostic and no timeout. Rewrite the OpenCode reviewer block to invoke `--format json` as the primary call and reconstruct the review from the assistant `text` parts (jq). Capture stderr to a `.err` sidecar (mirrors the Codex block). When the agent emits no text, surface the stop reason, output-token count, and stderr so the failure is diagnosable. Gate the stub on the extracted CONTENT, not the output file size — an empty jq extraction still prints a lone newline that a `[ -s file ]` check would treat as populated. Document the wall-clock timeout as a Bash-tool param (macOS lacks GNU timeout; opencode has no native timeout flag). review.md was already at the DEFAULT size-tier ceiling (40956/40960), so the fix cannot fit without reclassifying it into the LARGE tier (it is a multi-reviewer orchestration file that outgrew "focused single-purpose"; 43.4 KB sits well under the LARGE high-water mark). Recapture the 16 golden-install fixtures — the diff is exactly one review.md hash per runtime. Regression block folded into review-default-reviewers-workflow.test.cjs (new bug-NNNN test files are not accepted). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore(#1936): add changeset * test(#1936): property-test the OpenCode review jq reconstruction Address the re-review's one actionable finding: the jq JSON-event → text reconstruction had no fast-check property test. Add tests/opencode-review-reconstruction.property.test.cjs. It extracts the two shipped jq programs (OPENCODE_REVIEW, OPENCODE_DIAG) verbatim from gsd-core/workflows/review.md and runs the real jq — not a reimplementation — so the shipped logic is what gets tested. Properties: the reconstructed review equals the newline-join of every assistant text part (order preserved); a stream with no text part reconstructs to empty (drives the #1936 stub); null/absent text parts are dropped, never rendered as "null". Plus example-based coverage of the diagnostic edges the reviewer cited: missing .tokens.output and no step_finish degrade to "?"; non-JSON stdout makes jq fail rather than masquerade as a review. Verified the invariant has teeth (a comma-join jq fails the property). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(#1936): skip jq reconstruction property test when jq is absent The property test shells out to `jq`, which GitHub's windows-latest runners do not ship (macOS/Linux runners do). `execFileSync('jq')` therefore ENOENT-failed the whole file on `test (windows-latest, *)`. Probe `jq --version` at load and skip the suite when jq is not on PATH — the reconstruction logic is platform-independent, so the assertions still run in full on every jq-present runner (mirrors how golden-install-parity skips on win32). Verified: jq present → 7 pass; jq removed from PATH → 7 skipped, 0 fail. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(#1936): skip jq reconstruction property test on Windows, not just when jq is absent The prior guard skipped only when `jq` was absent from PATH — but the windows-latest runners DO ship jq, so the suite still ran there and failed with `jq: parse error: Invalid numeric literal` (confirmed from the CI job log). Root cause is Node's child_process argument quoting mangling the jq program (it embeds double quotes) on Windows, not the shipped review.md logic — the macOS/Linux legs pass. Gate the suite on `process.platform === 'win32'` (still also skipping when jq is absent), mirroring golden-install-parity's win32 skip. Logic is platform-independent and fully asserted on every macOS/Linux CI leg. Verified: macOS → 7 pass; simulated win32 → skips. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
Behruz Nassre Esfahani
|
2bada4de1a |
fix(#1698): capture codex review via --output-last-message, not stdout
The Codex reviewer in review.md captured the review by redirecting codex exec's stdout to the review file. On Windows, codex writes process-teardown output to stdout after the final agent message, so that noise was appended to a non-empty file and slipped past the `[ ! -s ]` empty-output guard as a silently polluted review (consumed by severity extraction and the plan-review-convergence gate). Capture the final message via codex's own `-o/--output-last-message <FILE>` and discard stdout. The #1115 contract is preserved (stderr to .err, capability-gated $CODEX_BYPASS_FLAG, --ephemeral, --skip-git-repo-check) and the empty-output fallback still fires when codex leaves no/empty output. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
|
Tom Boucher
|
b0d5ca3379 |
feat(#1517): support custom reviewer instances for /gsd:review (#1766)
* feat(#1517): support custom reviewer instances for /gsd:review Add a bounded review.reviewer_instances config surface so one model-capable adapter (e.g. opencode) can run as several independent reviewer identities in a single /gsd:review pass. Instances participate only via review.default_reviewers, expand before built-in slugs, are available iff their cli is detected, and a non-matching entry is a hard error (typo must be loud). >=2 same-cli instances emit a shared-adapter caveat in REVIEWS.md. Default path with no instances is byte-for-byte unchanged. Single-source instance->cli resolution lives in resolveReviewerSelection / normalizeReviewerInstances (parity-locked in tests/review-reviewer-instances.test.cjs). cli validated against KNOWN_REVIEWER_SLUGS only (never arbitrary shell); model/agent opaque, never shell-interpolated. Closes #1517 * chore(#1517): backfill changeset pr:1766 --------- Co-authored-by: review-bot <review-bot@gsd> |
||
|
Behruz Nassre Esfahani
|
ac40f070ef |
feat(#1318): require external reviewers to verify plan claims against source (#1421)
* feat(#1318): require external reviewers to verify plan claims against source /gsd-review built its external-reviewer prompt from plan text only and never asked reviewers to open the repo and verify claims, so a grounded HIGH could be outvoted by ungrounded LOWs. Add a concise, generic source-grounding block to build_prompt's Review Instructions: treat yourself as running in the working tree, open referenced files, cite path:line + mechanism, trace asserted mechanisms, downgrade to an open question if you have no file access, and know that grounded findings are weighted more heavily. Also clarify that CodeRabbit (a diff-only reviewer that never receives the prompt) must not be weighted as a grounded plan-level verdict in consensus synthesis. Workflow stays under its size cap (baseline bumped deliberately). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore(#1318): add changeset for reviewer source-grounding Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore(#1318): mark changeset docs-exempt (internal reviewer-prompt wording) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs(#1318): document reviewer source-grounding in COMMANDS.md; drop docs-exempt Review: a user-visible behavioral Changed warrants a docs touch, not a docs-exempt. Add a sentence to the /gsd-review entry in docs/COMMANDS.md (reviewers verify against source, cite file:line, grounded findings weighted higher) and remove the changeset docs-exempt marker so lint:docs passes via docs-updated. Also note the literal build_prompt test anchor. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(#1318): harden build_prompt fence extraction to be fence-run-aware Addresses maintainer review on PR #1421 (required-before-merge). The buildPromptReviewInstructions() test helper located the closing fence with `src.indexOf('\n```')`, which terminates at the FIRST triple-backtick line — so a build_prompt ```markdown block whose body embeds a fenced code example would truncate mid-content (dropping the `## Review Instructions` section) and give a spurious failure or false pass. Since this feature feeds source/plan content (which routinely contains code fences) to reviewers, that is a live fragility. Rewrite the extraction to be fence-run-aware, mirroring the CommonMark close rule in src/markdown-sectionizer.cts stripFencedCode: parse the opener's backtick run length, then close on the first line with >= that many backticks and only trailing whitespace — so a shorter nested fence is treated as content. Add a fail-first regression test (a 4-backtick outer fence wrapping a nested ```bash block) asserting the trailing `## Review Instructions` still extracts. Test-only change; no production .cts touched. Verified: test file 7/7, empirical fail-first proof the old indexOf logic truncated, full suite 4236/4236, eslint clean. Codex review: approve. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Co-authored-by: Tom Boucher <trekkie@nomorestars.com> |
||
|
Tom Boucher
|
fc37ae0c4f |
fix(#1115): capability-probe codex hook-trust bypass flag in /gsd:review; fail loud on empty output (#1122)
On codex-cli < 0.137.0 the review.md `codex exec` invocation passed --dangerously-bypass-hook-trust (added in 0.137) unconditionally and discarded stderr, so codex exited "unexpected argument" before reading the prompt and the empty output file was treated as a completed review — a silent degraded review. - Capability-probe the flag (`codex exec --help | grep`) and apply it via $CODEX_BYPASS_FLAG only when supported (works fine without it on older CLIs). - Capture codex stderr to a .err file instead of /dev/null, and replace an empty output with a diagnostic so a broken reviewer is surfaced (mirrors Cursor/OpenCode). - Update enh-773 enforcement test to require the capability gate + fail-loud guard instead of the unconditional flag. Closes #1115 Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
7c07fce70f |
fix(#381): make gsd_run launcher reachable in fresh-shell-per-block runtimes (#1084)
* fix(#381): make gsd_run launcher reachable in fresh-shell-per-block runtimes On runtimes that execute each fenced bash block in a separate shell process (e.g. Claude Code — documented behavior: each Bash command is a separate process; inline shell functions and exported vars do not persist between calls), the once-per-file gsd_run() function was undefined in every block after the preamble block, and the call was swallowed by `2>/dev/null || echo "{}"` into silent empty state. Fix (budget-neutral session-level resolution): - Ship gsd-core/bin/gsd_run, a POSIX sh wrapper that symlink-resolves its own location and execs the co-located gsd-tools.cjs. Exposed on PATH via the npm `bin` field (global installs) and shipped to local installs via the recursive gsd-core/ copy. - The per-file launcher preamble now appends `export PATH='<bindir>':"$PATH"` to the file named by $CLAUDE_ENV_FILE (Claude Code's documented env-persistence mechanism) so later fresh-shell blocks resolve gsd_run from PATH. Guarded as a strict no-op when CLAUDE_ENV_FILE is unset; the inline gsd_run() definition remains the fallback for all other runtimes. The single-quoted dir neutralizes shell metacharacters at source time. - Propagated via scripts/sync-runtime-launcher.cjs to all launcher-using files. - XL workflow byte budget 93000 -> 93200 (the ~130B clause pushes plan-phase.md to 93135; legitimate content growth, ratchet-up per #717). Regression tests (I)/(J) in runtime-launcher-parity.test.cjs cover wrapper delegation and end-to-end PATH persistence (sourcing the env file with a space-bearing install path). Known limitation: an install path containing a literal single-quote yields a malformed env-file line and falls back to the status quo (no regression); rare on sanitized home directories. Closes #381 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * docs(#381): add changeset for gsd_run fresh-shell reachability fix Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * test(#381): scope test (J) bare-PATH execution to POSIX (Windows Git Bash exec bit) Windows Git Bash (msys2) does not honor Node's chmod exec bit for PATH-executing extension-less scripts, so the bare `gsd_run` command lookup failed there even though the env-file PATH persistence was correct. The env-file content assertions (the fix's actual cross-platform logic) still run on every platform; only the final source-and-execute sub-step is gated to non-win32. Global installs on Windows are covered by npm's generated bin shim. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Jeremy McSpadden
|
fb37fa7dd5 |
fix(#725): route Codex gsd-tools calls through shim (#731)
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
a90c654745 |
fix(#891): probe non-Claude runtime homes in gsd-tools launcher shim detection (#911)
- Updated `gsd-core/workflows/_runtime-launcher.snippet.sh` with 15 new `elif` arms covering Hermes, Cursor, Codex, Gemini, Copilot, Windsurf, Augment, Trae, Qwen, CodeBuddy, Cline, Grok, Antigravity, OpenCode, and Kilo (respecting each runtime's env-var override with a `$HOME`-relative default). - Re-ran `scripts/sync-runtime-launcher.cjs` to propagate the expanded snippet into all `gsd-core/workflows/*.md` files (~70 files). - Manually applied the same snippet update to `commands/gsd/import.md` (1 occurrence) and `commands/gsd/graphify.md` (5 occurrences) — these are not covered by the sync script. - Updated `tests/workflow-size-budget.test.cjs` budgets (XL/LARGE/DEFAULT + discuss-phase target) to account for the ~3 KB snippet expansion. - Added regression test `tests/bug-891-non-claude-runtime-home-fallback.test.cjs` (6 tests: structural probe presence, ordering, behavioral HERMES_HOME env-var + default-path stubs, resolution order, and workflow propagation). - Added `.changeset/891-launcher-non-claude-runtime-homes.md` (Fixed). Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
2860e995b5 |
feat(#773): add --ephemeral and --dangerously-bypass-hook-trust to automated codex exec wrappers (#824)
* feat(#773): add --ephemeral and --dangerously-bypass-hook-trust to automated codex exec invocations Automated codex exec calls in the review workflow now carry --ephemeral (no session-state accumulation across CI runs) and --dangerously-bypass-hook-trust (skip hook-trust prompts for hooks whose provenance gsd-core already controls). Both flags were verified present in the installed codex CLI (codex exec --help). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#773): correct changeset pr: reference to #824 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
ecc42cefc6 |
fix(#669): /gsd-review --cursor actually invokes cursor-agent (#686)
* fix(#669): /gsd-review --cursor actually invokes cursor-agent The Cursor reviewer branch in review.md never ran the agent: - detection probed `cursor` (the IDE launcher) instead of the headless `cursor-agent` binary - the invocation used the two-token `cursor agent` (the IDE treats `agent` as a file-path argument, so the agent never starts) - the prompt was piped via stdin, but `cursor-agent -p` reads the prompt from a command-line argument, and `2>/dev/null` hid the empty result Probe `cursor-agent`; invoke `cursor-agent -p --mode ask --trust --output-format text` with the prompt passed as a file-path-reference argument (avoids the OS arg-length limit on large prompts); capture stderr so failures are diagnosable. Invert tests/cursor-reviewer.test.cjs to assert the corrected contract, with negative guards against the two-token form and the stdin pipe. The sibling `agy` reviewer already used the argument form. Closes #669 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * chore(#669): set changeset pr number to 686 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
afe61f015c |
fix(#687): bound agy print mode with its native --print-timeout (#689)
`/gsd-review --agy` hung indefinitely on large prompts. agy's print mode runs the full tool-enabled agent, and on a big, file-path-rich prompt its agentic Cascade loops on the code_search/grep tool and never converges; the transcript fallback only runs after agy exits, so it can't recover a run that never exits. The agy CLI exposes no per-tool deny (that lives in the Antigravity SDK), but it does expose --print-timeout — agy's native print-mode cap. Pass it explicitly so a stalled run self-terminates through the tool's own mechanism; a non-zero exit discards any partial output so the existing transcript fallback / "review failed" stub take over. Adds a regression test. Closes #687 Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |
||
|
Tom Boucher
|
463cffd894 |
chore(#604): rename get-shit-done/ runtime directory to gsd-core/ (#615)
* chore(#604): rename get-shit-done/ runtime directory to gsd-core/ Renames the installed runtime directory `get-shit-done/` to `gsd-core/` so the on-disk name matches the package (`@opengsd/gsd-core`), repo, and binary (`gsd-tools`). The npm package name and binary are unchanged; npx/npm consumers are unaffected. Mechanical (bulk, ~90% of the diff): - `git mv get-shit-done gsd-core` - Swept path/identifier references across the repo via `perl -pe 's/get-shit-done(?!-\w)/gsd-core/g'`. The negative lookahead preserves the five legitimate slug variants that are NOT the directory: get-shit-done-{OLD,cc,classic,cli,redux} (old package/repo names). - Build/manifest wiring: package.json (bin, files, coverage globs), tsconfig.build.json (outDir), ~86 .gitignore build-output entries, stryker.config.mjs, scan-ignore files, install.js path strings. - Frozen (not rewritten): CHANGELOG.md history; translated docs (README.<locale>.md and docs/{ja-JP,ko-KR,pt-BR,zh-CN}/). New logic (review here): - src/installer-migrations/003-rename-get-shit-done-to-gsd-core.cts: a proper ADR-0008 installer migration. On upgrade it walks the legacy `~/.claude/get-shit-done/` tree, classifies each file via the prior install manifest, and emits remove-managed / backup-and-remove for managed files while PRESERVING unknown user-added files. Symlink-safe (skips a symlinked root and symlinked entries; bounds-checks every path under configDir). The framework rolls back on install failure. Emptied dirs may remain (framework has no recursive dir-removal primitive) — documented. - scripts/lint-legacy-dir-name.cjs: CI regression guard forbidding the bare `get-shit-done` directory token (split token to avoid self-match; case- insensitive; `(?!-\w)` lookahead allows the slug variants; allowlists CHANGELOG, translated docs, and `gsd-allow-legacy-name` marker lines). Wired into the lint-tests CI job. - Restored scripts/lint-package-identity-drift.cjs detection regexes (the mechanical sweep had wrongly rewritten the old-name patterns it exists to detect) and marked them as intentional legacy references. - TDD tests for the migration and the guard; do.md slash-command guard regex tightened so a `/gsd-core/bin` path segment is not mistaken for a command; changeset + docs/installer-migrations.md row added. Breaking: the installed runtime path moves `~/.claude/get-shit-done/` -> `~/.claude/gsd-core/`. Migration 003 removes the stale legacy dir's managed files (preserving user files) on upgrade. Users with custom hooks/configs hardcoding the old path must update them. Closes #604 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): unsweep pending changesets + allowlist injection-example docs CI fixes for the rename PR: - Do not sweep pending .changeset/*.md (ephemeral release-note fragments, like CHANGELOG); reverted those body edits so 5 pre-existing malformed fragments (missing type/pr) no longer enter the PR diff and trip docs-lint. Allowlisted .changeset/ in the legacy-name guard accordingly. - Allowlisted TEST-EXAMPLES.md and docs/explanation/security-model.md in prompt-injection-scan.sh: they contain intentional injection examples / security-model prose; the path-reference rewrites are kept. CodeQL alerts on this PR are pre-existing (alert lines unchanged by this PR; none in the new migration/guard) and are out of scope for the rename. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): resolve CodeQL alerts surfaced on this PR The rename diff touched files carrying pre-existing CodeQL findings; per the no-pre-existing-dismissal rule, fixing every surfaced alert rather than waving them off. All behavior-preserving: - scripts/ci-test-scope.cjs: build the config-path match from string .includes() instead of a RegExp over an arg-derived value (js/regex-injection). - src/profile-output.cts: escape backslashes before pipe-escaping desc/safeName so the table-cell escape is complete (js/incomplete-sanitization). - tests/{bug-2643,bug-2808,docs-parity-live-registry}: two-pass HTML-comment strip so a bare/unclosed `<!--` cannot survive (js/incomplete-multi-character-sanitization). - tests/inline-plan-threshold: drop the no-op `\s`->`\s` identity replace, keep the meaningful POSIX-class conversion (js/identity-replacement). Verified: build:lib green; the touched test files + ci-test-scope + profile-output suites pass; lint:legacy-name clean. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): correctly resolve remaining CodeQL alerts (regex-injection + sanitization) The prior commit's fixes for two alerts were ineffective: - ci-test-scope.cjs js/regex-injection: the alert is the CLI-arg-derived `file` reaching static regex `.test(file)` calls (not the config rule). Removed ALL regex over file/t — startsWith/includes/=== string checks + an isWindowsHint helper — so there is no regex sink for the tainted value. - js/incomplete-multi-character-sanitization (3 test files): a single `.replace(/<!--...-->/g,'')` can let `<!--` re-form. Replaced with a fixpoint loop (replace until stable) plus a final bare-opener strip. Verified: no regex over file/t remains; ci-test-scope + the 3 test suites pass; lint:legacy-name clean. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): make ci-test-scope + comment-strippers regex-free to clear CodeQL CodeQL flags the regex PATTERNS syntactically (regex-injection on the --files arg split; incomplete-multi-character-sanitization on the <!--...--> replace), so loop fixes do not satisfy it. Made these paths regex-free: - ci-test-scope.cjs splitFiles: char-by-char separator tokenizer (no /[,\\s]+/). - 3 test files: indexOf/slice HTML-comment stripper (no .replace(/<!--/)). Behavior preserved; ci-test-scope + the 3 suites pass; guard clean. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): unblock security base64 scan on the large rename diff The security job hit its 10m timeout: base64-scan.sh choked on the binary test fixture tests/feat-3594-parser-property-style.test.cjs (embedded NUL/ non-UTF8 bytes -> thousands of bogus blobs + "ignored null byte" warnings), and the ~800-file rename diff is slow to scan regardless. - scripts/base64-scan.sh: skip binary-by-content files (grep -Iq .) — they can't carry base64-obfuscated *text* and feeding NUL bytes through the per-line scanner is pathologically slow. collect_files already filtered binary *extensions*; this catches binary *content* in text extensions. - .github/workflows/security-scan.yml: raise the security job timeout 10m->30m to accommodate very large diffs (the scan itself is unchanged). Verified locally: scan skips the fixture, 0 "ignored null byte" warnings, 0 findings, exit 0. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): sweep get-shit-done refs introduced by merging next The branch was updated with next (#614/#384/#618 etc.), which reference the get-shit-done/ dir (still named that on next). Swept the stale references in the merged files to gsd-core so the rename stays consistent and lint:legacy-name passes: - commands/gsd/discuss-phase.md (runtime-launcher shim paths) - src/core.cts (getAgentsDir layout comments) - tests/bug-384-agents-runtime-aware.test.cjs (require path to runtime lib) Verified: guard 0 violations; build green. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): exclude gsd-core/ path segments from bug-3683 command cross-ref invariant The #614 runtime-launcher shim added to discuss-phase.md references `${_GSD_RUNTIME_ROOT}/gsd-core/bin/...`. bug-3683's REF_PATTERN excluded path-y refs only via lookbehind, but `}` precedes `/gsd-core/` in the shim, so it mis-read the directory path as a dangling `/gsd-core` command ref (same class as the #604 bug-2954 fix). Added a trailing `(?![\w-]*\/)` so `/gsd-<x>/...` path segments are not treated as slash-command references. Verified locally on BOTH platforms before pushing: - mac (node 26) full suite: 0 failures - gsd-test-runner (linux, node22 image) full suite: 0 failures - bug-3683 + bug-2954 pass. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): lazily resolve findProjectRoot in gsd-tools (harden flaky CI) CI intermittently failed state.test's gsd-tools subprocess with "findProjectRoot is not a function" (flip-flopping across legs; not reproducible on mac full suite, gsd-test linux full suite, test:unit, or state.test x8). findProjectRoot is a re-export from core.cjs (sourced from project-root.cjs); binding it via destructure at module-load can be undefined under a load-ordering edge. Resolve it lazily at call time via a small wrapper so the lookup happens after core.cjs is fully initialized. Verified green on BOTH platforms before pushing: - mac (node 26) full suite: 0 failures - gsd-test-runner (linux, node22) full suite: 0 failures - state.test.cjs: 106/106; gsd-tools loads cleanly. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> * fix(#604): allowlist verification-patterns.md placeholder examples in secret scan The rename git-mv'd references/verification-patterns.md into gsd-core/, pulling it into the secret-scan diff. It documents stub/placeholder RED-FLAG env-var examples (illustrative Stripe test-key / database-URL / API-key placeholders) — not real credentials. Added it to .secretscanignore with the strict annotation, mirroring the existing gsd-core/workflows/plan-phase.md exception. Verified locally: secret-scan-lint --strict OK; secret-scan --diff origin/next exits 0 with 0 findings. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> |