Repository navigation
Conversation
…ons (GST-537) 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>
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…ude Code 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>
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
|
Merging to
After your PR is submitted to the merge queue, this comment will be automatically updated with its status. If the PR fails, failure details will also be posted here |
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
Merged
4 tasks done
4 tasks done
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> # Conflicts: # test/eng-finding-retry-budget.test.ts
Contributor
|
Frozen in wave |
Closed
5 tasks done
capy-ai Bot
added a commit
that referenced
this pull request
Nov 2, 2026
* fix(eval-plan): per-arm pilot projection and --seat-metered arms (#3653) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): guard tolerates Claude Code's cwd-reset footer and recovers after an unreadable init (#3651) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: pre-landing review fixes (journal reader strips the cwd-reset footer; an init with no text block no longer locks the session) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: journal reader skips a failed or unreadable init, as invocation() does Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * feat(autoplan): --ceo-only runs the CEO review alone with zero questions (GST-537) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test(eval-plan): all-seat pilot, single-attempt arm and --seat-metered CLI wiring Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(eval-plan): summed deviation adds arm deviations, matching the summed band (arms share rows) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: pre-landing review fixes (test pins an unpriced seat-metered arm) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: pre-landing review fixes (seat-metered declaration fixed per --out, price row still required, a real charge settles as spent) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: pre-landing review fixes Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test: classify autoplan-ceo-only in the Codex host metadata Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: a seat-metered arm that settles a charge is projected as a priced arm Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: select autoplan-ceo-only on the skills it runs; say it needs Claude Code Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(skills): /eval-plan --seat-metered and per-arm projection Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(eval-plan): skill prose stays under its context budget and carries no $N token Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test: run autoplan-ceo-only in the marathon lane Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(skills): a seat-metered arm leaves the seat set once it settles a charge Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs: changelog entry for /autoplan --ceo-only Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(autoplan): init_failed covers unreadable init output and the cwd-reset line (#3651) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test: count the autoplan-ceo-only paid file in the shard inventory Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test(autoplan): #3651 tests pass the test-code type ratchet Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(dx): CAP finding titles carry a one-line description (GBRA-223 claim 2) The DX error contract told both autoplan DX voices to title capped findings exactly CAP4:<path>:<item>, so gates, tasks and reconcile saw a bare code. The title is now CAP4:<path>:<item> <what fails>; the grader already parses that shape. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): reconcile pairs same-file findings that carry no line (GBRA-223 claim 1) Plan reviews name files, rarely lines, and reconcile associated by location only when both findings had a line, so findings both voices raised on one file were filed new and native-only. A same-file pair with a missing line is now associated as unverified (basis file), never confirmed; a CAP code no longer counts in title comparison. DRIVER.md names submit --reconcile. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): open Critical and High findings reach the gate as apply items (GBRA-223 claim 3) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): a spend cap binds on every unattended host; --estimate-usd persists; DRIVER names --usage-usd (GBRA-223 claim 4) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): run binds complete api results, names a truncated response, --max-tokens; DRIVER documents run (GBRA-223 claim 6) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * feat(autoplan): carry an item-by-item answer when the later gate asks nothing new (GBRA-223 claim 7) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(dx): trim the compact error contract so plan-devex-review stays under its size pin Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: pre-landing review fixes (p1b applies finding fixes, api attempts reserve no host default, cross-file title pairing, paired severity counted) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: review fixes (priced api reservations, no file-only twins, two-pass pairing, partial replies keep p1-dependent defaults open, outside refusal degrades) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * P7: record whose words answered a gate (--words-by agent:<id>), never carry an agent's own words, import each row under its own answerer (GEVA-2) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix: review pass 2 (p1b needs --rec-hash after recommend; api calls send the output limit they reserve for) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * changelog: fragment for the /autoplan field-report wave (GST-672) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs: captain record notes --words-by agent attribution Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * test: follow finding apply items in approval-class and Captain flow tests; bind run-api limit to DEFAULT_MAX_TOKENS Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * wave autoplan: lower the gemini plan-devex-review pin to its exact size after #3781's DX trim Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): CAP titles on different paths never match by title in reconcile or confirm rounds (wave review) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): a truncated api call settles its cost and a retry is admitted afresh against the cap Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): an applied finding stays open until the revised plan is reviewed; a shared location alone never twins two findings Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(approval): a plan gate answered in an agent's own words is not owner approval Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(gate): a new answer replaces the earlier answer's authorship and words Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): a failed api call with no usage keeps its attempt open; reservations price the bound prompt (wave fix-check) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): a spend cap fails closed: only a call refused before sending keeps its attempt; a sent call with unknown usage settles at its reservation Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(autoplan): the 16384 output default applies to both providers; changelog for the wave's integrator fixes Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(autoplan): how run settles a failed api call; changelog scopes it to the native voice Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * fix(autoplan): a spend cap fails closed on every exit of run; an unpriced model reserves the default once a cap is set Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(autoplan): every sent call settles; an unpriced model under a cap reserves the default Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * docs(autoplan): DRIVER and help say a cap reserves in any session kind and how run settles a sent call 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>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed
New
/autoplan --ceo-onlysection (autoplan/sections/ceo-only.md): it runs the CEO phase only, asks no questions, endsstatus=reviewedand approves nothing.gstack-ceo-mode-handoff --ceo-onlynow names Step 0E's rule. Before, it failed on an unknown flag.Text before/after
Why
Garry wanted a CEO review that "just automatically does the right thing", and the planning eval (GST-538, due Nov 3) needs a gstack CEO arm that a harness can run without a simulated user. Before this change, /autoplan always went on to the Eng review and stopped at an approval question, and standalone /plan-ceo-review asks for the mode and offers to chain. Now
/autoplan --ceo-only --no-outsideruns the CEO review alone, takes every recommendation, and leaves aceo-review.mdreport.Summary
/autoplan --ceo-onlyruns Phase 0, the outside-reviewer preflight and the CEO phase. Every question takes its recommendation and gets an audit-trail row. User challenges are listed and never applied. The run endsGSTACK_RESULT: skill=autoplan status=reviewed run=<dir>with<run dir>/ceo-review.md.--no-outsidekeeps the whole review on one model.autoplan/SKILL.md, and a free test pins that no other autoplan file reads it, so a normal /autoplan run never loads it.status=reviewedjoins theGSTACK_RESULTstatuses.docs/unattended.mdgets a "CEO-only runs" section. It says the mode needs Claude Code for now, and thatgstack-autoplan(the driver) andrun.jsonget it in PR 2.autoplan-ceo-onlyfails if the run asks anything, loads a skipped phase's file, misses aceo-review.mdheading, ends with another status, or leaves new files in HOME outside its folders.Plan: GST-537 gate brief (captain drive
plans/GSTA-215.md), autoplan run autoplan-20261031180042, approved with "Do it".Blast radius
The prose change is one new section file plus one section-index row in autoplan's Claude render. The section is read only when the invocation names
--ceo-only, andtest/autoplan-ceo-only.test.tschecks that no other autoplan section, the driver orlib/autoplan-*.tsnames it (script ran).ResultStatuswidens by one value thatparseResultaccepts, and nothing inrun.jsonwrites it yet (file read:lib/headless-artifacts.ts). Non-Claude hosts inline autoplan and do not carry the section, so--ceo-onlydoes nothing there. The docs say so.Gate changes
test/helpers/carve-guards.ts: autoplanmaxSizeRatio1.1511 → 1.1723. That is the measured 119,545 / 101,979, which is +2,276 B for the section and its index row (GST-537, GStack Minder approved A1, 2026-10-31). A free test pins that a normal run never reads the section.autoplan-ceo-onlyruns in the marathon lane (weekly, non-blocking). Autoplan PRs no longer run this case before merge. It measures 349–984 s, and gate shards must stay under 600 s (I5,test/paid-shards.test.ts). GStack Minder approved moving it to the periodic tier, but the periodic lane is at its 40-slice cap, so it sits in marathon (7 of 8 slots), likeplan-ceo-split-overflowanddesign-motion-showcase. No workflow file changed.Behavior changes
reviewedinGSTACK_RESULTlines (lib/headless-artifacts.ts).gstack-ceo-mode-handoffaccepts--ceo-only;--autowith--ceo-onlyexits 2 ("--auto and --ceo-only conflict").Approvals
Source: gstack-approvals record for gsta215-ceo-only-skill (display only; not an approval citation).
Test Coverage
All new code paths have test coverage: each of the 6 changed runtime files has a changed test that names it (deterministic coverage map).
Tests ran: typecheck count unavailable (exit 0)
Tests ran: ceo-only 208
Pre-Landing Review
No issues at the blocking floor. Pass 1 (specialists, then the red team) found 2 informational test gaps, both fixed: the skipped-phase check now reads Bash, Grep and Glob arguments, and the test's temp dirs are cleaned up. Four fix-checks found nothing. The final one ran on the frozen tree. A native adversarial pass on the frozen tree found nothing at the floor.
REVIEW_LOOP: passes=1 blocking_found=0 deferred=0 noted=1 fixed=2 scope=frozen fix_checks=4 dispatches=4
Gate edits: 1 listed, 1 read, 1 finding — 1 kept (reason below), 0 restored, 0 open
test/helpers/carve-guards.ts, kept: GStack Minder approved A1 on the measured ratio (GST-537, 2026-10-31)Outside review (openai/gpt-6.1-sol): request changes on two Medium items, both below the floor. Non-Claude hosts don't carry the section: documented as Claude Code only.
ceo-review.mdlives in the shared.gstack/tmp/autoplandir, so a later run in the same repo replaces it: noted, and PR 2's driver gives each run its own folder.Review notes
Below the blocking floor (advisory, informational, test-gap or wording items); recorded here and not fixed in this PR.
lib/headless-artifacts.ts:206:reviewedis inResultStatusbut not in the run.json schema or the terminal sets (PR 2 adds it with the driver flag).autoplan/sections/ceo-only.md.tmpl:7: in aSESSION_KIND: unattendedsession, both this section andsections/unattended.mdapply (PR 2's driver flag covers parent-driven runs).autoplan/SKILL.md:269: the section-index row comes after the prerequisite offer, so zero questions relies on the model reading ahead. All 5 paid sessions here asked nothing.autoplan/sections/ceo-only.md.tmpl:10: taking a preamble block's non-consenting default may record a decline on a first run.autoplan/sections/manifest.json:49: non-Claude hosts inline autoplan and do not carry the section.autoplan/sections/ceo-only.md.tmpl:31:ceo-review.mdlives in the fixed.gstack/tmp/autoplanfolder.test/skill-e2e-autoplan-ceo-only.test.ts:99: the stray-write check sees new files under HOME only. The changelog wording now says so.Eval Results
autoplan-ceo-only(run gsta215-e2e-4, branch tree before the review fixes): PASS, $4.75, 62 turns, 860 s. Zero questions,status=reviewed,ceo-review.mdcomplete, no stray writes. Earlier runs: run 2 was red only on Bun's cache under HOME, and run 3 only on the machine's git-wrapper log. Both were detector fixes.claude-opus-5-5): the planted CEO-level flaw was caught by--ceo-only3/3, standalone /plan-ceo-review 3/3, and the no-gstack arm 3/3. The blind judge (gpt-6.1-sol) scored all 3 decoys as misses.--ceo-onlyasked 0 questions. That is no detected loss at n=3. These home-field plans don't separate gstack from a plain review; GST-538 is the powered reading. Smoke cost: $21.54.Plan Completion
PR 1 items 1–5 of the gate brief: the skill path, the zero-question test, the no-loss smoke, docs, and the gate-change note. Items 6–10 (the
gstack-autoplan next --ceo-onlydriver flag) are PR 2.TODOS
No TODO items completed in this PR.
Documentation
Status: current. No documentation edits were needed.
Scope: base 2d1b55d to head ce100cb (14 changed files, nothing staged, unstaged or untracked) in
gsta215-ceo-only-skill. Authored docs, templates and the changelog fragment were audited against the code diff (lib/headless-artifacts.ts,bin/gstack-ceo-mode-handoff,autoplan/sections/ceo-only.md.tmpl).Documentation health:
docs/unattended.md: current. Thereviewedstatus, the CEO-only runs section and its phase scope match the code.autoplan/SKILL.md,autoplan/sections/ceo-only.md,autoplan/sections/ceo-only.md.tmpl: current.changelog.d/gsta215-ceo-only-skill.md: current, not modified.AGENTS.md,README.md,docs/skills.md: not stale, but do not mention--ceo-only.Coverage debt:
--ceo-onlyand--no-outsideexists only indocs/unattended.md.docs/skills.md/autoplandoes not list them (common gap).Test plan
bun run typecheck: exit 0bun test test/autoplan-ceo-only.test.ts test/codex-host.test.ts test/touchfiles.test.ts test/paid-shards.test.ts test/evals-workflow-wiring.test.ts test/parity-suite.test.ts test/skill-md-size-guard.test.ts: 208 passautoplan-ceo-only(gsta215-e2e-4): PASS