Garry Tanandcapy-ai[bot]
21277291c8
* feat(ship): GSTA-167 turn levers on the restored reviewers: ship preflight/release bins, own-jam meter, dispatch-pair metering, ship/DRIVER.md
Ports the non-reviewer parts of #3633 (gsta-167-pr34) and #3637 (gsta-167-pr5 @2914400)
onto main ae3d58c (v1.91.331.0). Reviewer briefs, checklists, the fix-check prompt and
the depth rules are byte-identical to main and now pinned in
test/review-packets-removed.test.ts.
Left out: gstack-review-packet and its libs, parallel/batched fanout, --like, collect,
the audit prompt split, the docs-audit packet, review-army/adversarial/outside-voice
edits, the gate-scan summary-only block and the review-depth scope-error change.
Packet-bytes reader pricing is dropped from gstack-ship-budget; handoff READ ranges
send each section whole; the driver runs the audit, review, adversarial and
documentation sections as written.
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* test(context-budget): raise the capy/skill/gstack eager ceiling to its captured value after the main merge (own-jam meter, driver and reviewer-batch prose, GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* test(ship-driver): name ship/DRIVER.md.tmpl so the pregate maps the template to its test lane (GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(ship-driver): keep the skill generator off lib/ship-budget so a minimal setup checkout still renders (GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* test(windows): keep test/ship-release.test.ts off the Windows subset; it launches the shebang bin and sh stubs directly (GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* feat(ship): variant E orchestration: one program per review pass and Step 14.5 audit, the review child starts at Step 9, a metered dispatch:docs role, and the driver's record headings (GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* feat(ship): variant G orchestration: one call per driver phase (9 for a clean pass, at most 12), gstack-ship-preflight --audit for single reads and deterministic audits, an observe-mode ledger with note and credits, 240 s wait pings, and small-model audit readers (GSTA-167)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(ship-budget): a Capy handoff child needs no jam id: verify-handoff with no reading makes it parent-metered (METER: parent), it never delegates on an unknown charge, and reconcile charges the parent's counter delta (GSTA-167 replay 7e)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(ship): a poll settles a reader only at done, failed or a final summary, never a fresh idle; return settles the child's open step reservation (GSTA-167 replay 7e)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(ship-budget): return releases an open step reservation under the estimate meter instead of booking its projection; measured meters still settle it (GSTA-167 replay 7e)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(review): a clean fix-check after below-floor fixes is done; Step 11 and fix-check below-floor findings are notes at pass 1
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* feat(ship): verbatim reviewer briefs and loss-free collection (gstack-review-dispatch render/sent/collect/judge/verify)
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix: pre-landing review fixes
- verify requires each output's brief rendered and sent verbatim, reruns judge
against the recorded judged.jsonl hash, and flags outputs collected after judge
- judge drops self-folds, fold cycles and chains
- prose findings parse a leading label or a label heading, not any severity word
- releasePaths: manifests count as release metadata only for version-only edits,
lockfiles never; TODOS.md stays out of lane-evidence exemptions
- push stops when the sealed-output guard did not run
- ship-audit reads the test tree once instead of one git grep per file
- drop the uncalled --delegate measured mode; doc/help step-number fixes
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix: pre-landing review fixes (pass 2)
- collect keeps a finding wherever its label stands (headings, table rows,
mid-line, Informational and upper-case labels); only top-level list items
inherit a label heading, and any heading or bold-only line ends it
- judge treats a merged file missing any collected id as a drop; verify flags
merged_ids_missing and a sent reader never collected (NOT_COLLECTED)
- finish computes release paths after --build, so a build that rewrites
package.json is a build-input change
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(review-dispatch): verify per render, voided superseded mismatches
- verify treats each render as one dispatch of its reader, so a two-pass
ledger compares pass-1 sends with pass-1 briefs (later renders print as
<reader>@2); outputs bind to the latest send of their own render
- a mismatched send re-sent byte-exact before anything was collected is void
and counted (voided=N on the verdict line); one that produced an output
stays MISMATCH
- an indented bold sub-line no longer ends a label section (fix-check)
- docs: a failed reader is collected from its failure text
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(review-dispatch): only the newest render of a path checks its file; absolute ledger paths
- a later render into the same briefs directory overwrites the earlier brief by
design, so verify checks a brief file only for the newest render of that path
- collect and judge record absolute output, merged and judged paths, so verify
reads them from any directory
- docs: a failed reader's failure text is parsed like any output
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* chore: changelog and TODOS
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* fix(ship-release): the changelog_fragments fragment is release metadata
Fix-check 4 (CRITICAL) after the main merge: releasePaths omitted
<changelog_fragments.dir>/<branch-slug>.md, so finish stopped on the
fragment Step 13 writes under gstack's policy, or routed it to re-review.
version now names the fragment in its Step 13 NEXT line instead of
[Unreleased], DRIVER.md Step 13 writes and lints where NEXT says, and the
preflight POLICY line says Steps 12-16.
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
* chore: changelog and TODOS
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
---------
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
12 KiB
/ship driver: run /ship from a parent agent
Read this instead of ship/SKILL.md when a parent agent on any host runs /ship
(SESSION_KIND: unattended). Commands live in <root>/bin/ (<root> holds ship/); run each with
the GSTACK_LAUNCH assignments your installer printed (. <the GSTACK_LAUNCH_FILE it printed>), never
export $(...), env, printenv or a bare set in a tool fence.
Rules. Each phase below is one call: one program (Capy: an execute program; else a shell
script) that runs the bins, reads, dispatches and waits, returning only result lines and reader
outputs; print them verbatim. The model keeps only lead judgment, fixes, CHANGELOG and PR prose, the
gates and the report. STOP: (exit 3) means nothing after it ran; NEXT: names
the next act. Read a section only when a phase or the list at the end sends you, never
up front. A STOP or decision this file does not cover: read ship/SKILL.md at that step, finish it
there, come back. Unattended, a question with a recommended option takes it (log it); a gate
disposition and every approval stay pending and end the run. Keep a private invocation record (the
--record file): write its six headings at Phase 1, then fill each as its phase ends: Checks
(Phases 2-4: lanes, pregate, evals, Coverage:, the plan checks, Scope Check:), Decisions (any
phase: each answered or auto-taken question, in its words), Reviews (Phases 3-5: the plan line,
start tokens, Pre-Landing Review:, Gate edits:, the reviewed tree), Release (Phases 5-7:
version, CHANGELOG, commits), Documentation (Step 14.5) and Next steps (every phase: the next act).
Reviewers are the sections' own. Every audit, specialist, Red Team, QA, fix-check, adversarial,
outside-voice and documentation brief, its severity and confidence rules and its gates come from its
section verbatim, with only its placeholders filled. This file restates none of them and
never shortens a reviewer's brief. A blocking reader's prompt is its gstack-review-dispatch render
file, unchanged; its findings reach merged.jsonl only via collect; judge refuses a drop or downgrade. Audit readers (Steps 7, 8, 9.2.1, 14.5) cannot block, so each runs on a
small model at low effort; their findings go to the PR body.
Calls. One pass with no fixes is 9 calls: Phases 1-8 and the report; at most 12 with a STOP's section read or a fix-check. A pass that fixes something repeats Phases 3-4.
Budget: observe. gstack-ship-budget start --mode observe records, never admits: each program ends
with gstack-ship-budget note --step <last step> --reading R --credits C, R and C read in that program
(Capy: your own jam's cost.tokens and cost.credits). No check per step, no reservation, no meter-only
call, no handoff; under --token-cap a note past the cap prints verdict=stop, the one stop check.
A dispatching program takes BEFORE (R and C) as its readers start and passes
--dispatched-at <R> --dispatched-credits <C> --readers <n> with the reading after the last settles.
Waits. No call waits past 240 s: a program returns whatever still runs at 240 s; the next call resumes (Capy: wait 240 s on it). A small
waiting child saves nothing. Past 100 minutes, stop.
One program per pass. Each review pass, Step 11 and the Step 14.5 audit is one program, and the
parent makes no other call while it runs: (1) gstack-review-plan --base <base> with the pass's
flags (its DISPATCH: lines; Step 14.5 has one reader, docs); (2) gstack-review-log --start <skill> once per row the step logs (Step 14.5 takes none); (3) BEFORE; (4) one subagent per reader on its rendered file, then
sent (BRIEF_MISMATCH voids it);
(5) a poll until every reader settles (status done or failed, or a final summary; a fresh idle
has not run); (6) note with the reading pair; a later batch (the Red Team,
given the specialists' collected lines) repeats (3)-(6); (7) the collect: return only each reader's
final output and status, via collect. The next call applies the section's merge, judgment, fixes and persistence;
it never walks the dispatch steps call by call.
1. Load: Step 0
The install check and this file.
2. Preflight, tests, evals and reads: Steps 0.9, 1, 2, 3, 4, 5, 6, 6.5, 10
gstack-ship-budget start --record <rec> --mode observe [--token-cap <n>] --reading R --credits C, then
gstack-ship-preflight --base <base> --lane <label>=<command>... (the test commands AGENTS.md or
CLAUDE.md names): STATUS:, DIFF:, PR:, CARRY:, SCOPE:, MERGE:, DISTRIBUTION:,
POLICY:, RUNTIME_LINES:, PREGATE:, LANE:. CARRY: carried ends the run on its GSTACK_RESULT:
line. A lane passes on tests_ran: N with exit 0; tests_ran: 0 is ZERO-RUN, a failed lane; unknown
is count unavailable. In-branch failures stop the ship; after a repair rerun with --tests-only.
requires-remote needs gstack-pregate clear before Step 18. When prompts, templates, judges or eval
harness changed, start the project's eval selector and pre-merge command (detached past 30 min); else
record evals: none affected. Steps 6.5 and 10 run only if gstack-greptile-early detect finds
Greptile. Then gstack-ship-preflight --audit --base <base> --out <run dir> writes the diff once
(CONTEXT:) and the TODOS.md items naming touched paths (TODOS:), and prints the coverage map, the
plan candidates and READER: coverage=<needed|none> plan=<needed|none>. The same program reads
sections/review-army.md, sections/adversarial.md, sections/documentation.md, the todos and commits
files, and sections/test-coverage.md or sections/plan-completion.md only when its reader is needed.
Never read TODOS.md or the full diff into the parent. A PERSIST: hint never replaces Step 9.
3. Audits and review: Steps 7, 8, 9, 9.1, 9.2, 9.2.1
One program: the audit readers READER: names (Step 7 from sections/test-coverage.md, Step 8 from
sections/plan-completion.md, each run as written), Step 9.2.1's QA reader (that step of
sections/review-army.md as its brief) and the pass (above): Step 9's plan and start, Step 9.1's
specialists and Step 9.2's Red Team as written. Coverage: and Step 8's line come from --audit when no
reader ran.
4. Judgment: Steps 8.1, 8.2, 9.3, 9.4 and Step 11
Steps 8.1 and 8.2 (Scope Check:) on the todos and commits files, then the core checklist, Steps
9.3, 9.4 and the repeat decision as written (gstack-review-plan included;
JUDGMENT: ok before any fix), fixes and persistence in
one program. Prints Pre-Landing Review:, Gate edits:. NEXT: stop stops; full or
scoped reruns Step 5 (--tests-only) and Phases 3-4; done: the same program runs Step 11 from
sections/adversarial.md as written (above), only the voices the last full-pass plan lists (neither
prints ADVERSARIAL: skipped (review plan)).
5. Voices, bind and version: Steps 11, 11.5, 12
Judge the voices; missing outside coverage is unavailable, never clean; queued fixes rerun Phases
3-4. Then gstack-review-dispatch verify (VERIFY: clean) and gstack-review-read: the final Step 9 row and Step 11 native row (or CURRENT freshness
after a planned skip) agree on wtree, start_wtree, end_wtree; save that reviewed tree. Then
gstack-ship-release version --base <base> [--bump micro|patch] --why '<line>'; MINOR or MAJOR is a
question; VERSION: skipped (policy: ...): write no version.
6. Notes and documentation: Steps 13, 14, 14.5
Step 13: one user-facing entry from the commits file, where NEXT: says;
gstack-prose-lint <it> --base origin/<base> --gate until PROSE_LINT: ok. Step 14: TODOS.md deferrals and
completions, edited from the todos file. Step 14.5, every ship: read sections/documentation.md and run its audit as written, as
one program (above), then gstack-ship-receipt docs-caps --base <base>; print Documentation:.
One re-audit at most.
7. Release: Steps 15, 16, 17
Commit features in bisectable chunks, then gstack-ship-release finish --base <base> --reviewed-tree <t> --docs-post <post> --lane ... [--build <cmd>] --trailer '<t>': ROUTE:, DOCS:, EVIDENCE: (FRESH;
STALE or MISSING reruns once), PREGATE:; print them, the last Gate edits: and each tests_ran. Then
gstack-ship-release push --branch <branch> [--gate-open <n>]: REDACT_PREPUSH:, SEALED:, PUSH:
(pushed or ALREADY_PUSHED). Its pre-push offer STOP runs Step 17's guard block (HOOK_STATE:,
HOOKS_IN_GIT_DIR:, PREPUSH_PROMPTED:); unattended takes the recommended install. Never force-push.
8. Publish: Steps 18, 19, 19-C, 20, 21, 22
- Step 18: the open-PR lookup; a title file starting
v<NEW_VERSION>.UNATTENDED_SHIP:with no open PR at HEAD takes Step 18-H. - Step 19 body: the receipt block; with
CAPY_PR: on(Step 19-C) What changed, Why, Blast radius, Approvals (gstack-approvals render); Summary, Test Coverage, Pre-Landing Review (REVIEW_LOOP:,Gate edits:, deferred, notes), Exploratory QA, Plan Completion, Verification Results, Issues, TODOS, Documentation, Test plan.gstack-prose-lint <top> --sections 'Why,Summary,Blast radius' --gate;gstack-issue-links;gstack-review-scope summary --since "$REVIEW_SINCE" && gstack-ship-receipt write --pr <n> --base <base> --gate @<f> --pregate <f> --issues @<f> --review-loop @<f> --docs @<f> --body <body>;gstack-post pr-bodyandpr-title(a new PR via the host's PR tool; a draft while a gate finding isopen).SHIP_RECEIPT:opens your report. - Step 20: the ship metrics row via
gstack-review-log. - Step 22:
gstack-ship-budget close --reading R --credits C(BUDGET_FINAL:andBUDGET_TIMELINE:carry tokens and credits; put both in the report), thengstack-ship-receipt result --pr <n>:GSTACK_RESULT: skill=ship status=complete run=<dir>. Under a merge queue,gstack-ship-queue ready --pr <n> [--screen|--no-screen]at the pushed head; paste its output.
Where each STOP sends you (paths under <root>/ship/)
merge conflicts:SKILL.md#step-3-merge-the-base-branch-before-testsNEXT: carry route:SKILL.md#step-1-pre-flightDISTRIBUTION: candidates=:SKILL.md#step-2-distribution-pipeline-checkAPPLE:with a store ask:sections/apple-release.md#apple-app-store-releasepregate failorincomplete:sections/tests.md#step-5-run-tests-on-merged-code- a failed or zero-run lane:
sections/tests.md#test-failure-ownership-triage - a red eval or a failed free-suite shard:
sections/measure.md#measure-then-fix-loop-a-red-paid-eval-or-a-failed-free-suite-shard - Greptile found:
sections/greptile-early.md#step-65-early-greptile-pr-after-the-free-tests-pass - coverage below target, a failed audit:
sections/test-coverage.md#step-7-test-coverage-audit - plan items NOT DONE or UNVERIFIABLE:
sections/plan-completion.md#gate-logic NEXT: stop:sections/review-army.md#decide-whether-to-repeat-step-9- a
REVIEW_SCOPE_*refusal, the review-log row:sections/review-army.md#step-94-fix-first-and-persistence - a required probe not passed:
sections/review-army.md#step-921-exploratory-qa-before-fix-first - native adversarial incomplete, queued fixes:
sections/adversarial.md#finish-the-adversarial-phase - Step 11.5 mismatch:
SKILL.md#step-115-bind-the-reviews - a
versionSTOP:SKILL.md#step-12-version-bump-auto-decide Documentation: blockedor aDOCS_CAP:over:sections/documentation.md#blocked-recovery- finish: changed since review:
SKILL.md#2-choose-the-change-route - finish: documentation inputs changed:
SKILL.md#3-resolve-documentation-freshness - finish: build, lane or pregate STOP:
SKILL.md#4-verify-the-frozen-candidate - a push STOP:
SKILL.md#step-17-push - unattended, no open PR at HEAD:
sections/handoff.md#step-18-h-handoff-without-a-pr-automation-lanes - a receipt write or
gstack-postrefusal:sections/pr-body.md#compose-the-body-then-publish-through-gstack-post verdict=stoporDELEGATE::sections/budget-handoff.md#budget-and-context-handoff-read-when-a-cap-is-set-or-verdictdelegate-printsQUEUE_RECEIPT_STALE:../docs/ship-policy.md#ship-before-ready