Skip to content

docs(goal-conductor): worker debate Round 0 biggest-risk rows and owned claim checks - #33783

Merged
KiroCrewCaptain merged 1 commit into
mainfrom
docs/skill-critic-roles
Nov 9, 2026
Merged

KiroCrewCaptain merged 1 commit into
mainfrom
docs/skill-critic-roles

Conversation

@iamwhatever

@iamwhatever iamwhatever commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: Worker-debate critics raise the key problem even when it sits outside their role, and every claim the debate relies on has a named checker.

In a worker debate, critic roles are fixed up front. The owner worried this can miss the main point (outside every role) and that dropping roles makes critics overlap. On the 2026-11-02 debate, a role's Round 0 risks got no table rows, and a proposer's claim that "an existing part already does it" was never checked by anyone, so a wrong claim passed review and failed live.

Why it matters

A debate that misses its key risk, or lets a false claim through, produces a confident wrong answer. The protocol is only as good as what it forces critics to check.

Not a goal

What changed (motivation → approach → change)

Design record: the worker debate critic-roles (outcome stable, 2 rounds + wrap up, three critics: focus, overlap, cost). Its note and final proposal live in the Kiro Crew research directory (research/debate-critic-roles/note.md, final-A.md). RFC #33408 (worker debate claim checks, docs/request-for-change/rfc-worker-debate.md) is merged with status accepted (merge commit 76855bb, 2026-11-09); this PR implements its Option B as amended by that debate.

One file, src/kiro_crew/builtin_skills/goal-conductor/references/worker-debate.md:

  1. Role card: Guards is "one goal" per card.
  2. Round 0: each critic also names its single biggest risk regardless of role (may point at one of its own ids) and at most three outside-role risks. Only blocking risks and the biggest-risk entry get rows; the outside-role list is for the Conductor.
  3. After Round 0: the Conductor reads the lists side by side. An unowned biggest-risk or blocking risk widens the closest card's Finding line (Guards stays one goal), sent with the existing round-1 critic relay.
  4. Note check: one named critic checks the note keeper's copy (was: every critic).
  5. Cross-critic duplicates: keep the first id, rule the later one same as <id>; the proposer answers once and both critics verify.
  6. Option B claim checks: question-form objections marked Q: from a question bank, answered with evidence only; proof cells on "existing part does it" mechanism rows; in each step-b relay one line per checker naming the ids/rows it checks, never by their author; the checker asks "what if it is wrong on the path the proposal uses it for" and lists a proof checked: line; an unchecked assignment stays open under its original id (a proof cell's id is proposal letter + row, e.g. A-M3) and is reassigned; one-critic fallback.
  7. Step a: "response table (round 1: Round 0 rows only)".

The A-M3 id prefix is a wording fix the overlap critic asked for in its final file (two proposals could both have an M3).

Cost: +43/−6 lines of skill text, 0 new rounds, 0 new relays, 0 new work items.

Backwards compatibility

Compatible: skill text only; no code, config or stored format changes.

Removes nothing: the only narrowed duty is the note-copy check, moved from every critic to one named critic; the check itself stays.

Tests

Proof: not provable -- docs or config only

Ran locally: scripts/docs-lint.sh (pass), scripts/check_builtin_skill_scope.py (pass), test/test_conductor_skills_steer_corrections.py, test/test_goal_conductor_skill_grants.py, test/test_builtin_agent_skill_resources.py (44 passed).

Manual verification

N/A — skill text; the design was verified by the three debate critics in their final files.

Related Issues

Refs #33408

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

@iamwhatever
iamwhatever requested review from a team and KiroCrewCaptain as code owners November 9, 2026 07:26
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Nov 9, 2026
@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ⏭️ skipped

Revision d5c4df9201b9218a54fee550dce7a72b4ab9d01e touches no user-facing surface (no changes under website/ or committed screenshots), so the UX review was skipped. Does not block merge.

@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of d5c4df9201b9218a54fee550dce7a72b4ab9d01e — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: this is skill-text-only and puts into practice an accepted RFC; no test or code reads the wording it replaces.

Notes

  • The text says one critic is "named on its card" for the note-copy check, but the role-card field list (Role/Guards/Finding/Out of scope/Done) has no slot for that duty. Either add the line to the card template or say the Conductor names that critic in the seed, or the Conductor may forget to assign it.
  • docs/request-for-change/rfc-worker-messaging.md:279 still says critics (plural) compare the note keeper's copy. That will go stale once this lands.

[DESIGN-REVIEWED] d5c4df9

Description read

[DESCRIPTION-READ] 6ff30dcc024a6a093ceb572e70709c2bca657d8af59653817d6268a245d62ed7

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of d5c4df9201b9218a54fee550dce7a72b4ab9d01e and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] d5c4df9

False positive or not applicable? A repository writer can comment:
/ai-review override gpt d5c4df9201b9218a54fee550dce7a72b4ab9d01e false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of d5c4df9201b9218a54fee550dce7a72b4ab9d01e — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Every item puts into the skill text a rule that Phase 1 of the base-accepted RFC rfc-worker-debate.md lists. It adds no code, config or stored format.

Notes

  • Item 6 — undeclared: the RFC (§2.4) says a proof cell's id "is its mechanism row". The diff writes "its proposal letter and mechanism row, e.g. A-M3", and that deviation rests only on one critic's file outside this repo. Subtraction: none needed if the A-M3 form is kept; the deviation is just not in the RFC.

What this change ships

Inventory (7 items) — 7 justified

Intent: make worker-debate critics raise the key risk even when it falls outside their role, and give every proposer claim a named checker who did not write it. ADDITION to the skill protocol.

  1. The role card's Guards field is limited to one goal. — justified
  2. In Round 0, each critic names its biggest risk plus at most three risks outside its role. Blocking and biggest-risk entries get rows. — justified
  3. The Conductor reads the Round 0 lists side by side and widens the closest card's Finding line for an unowned risk. — justified
  4. One named critic checks the note keeper's copy, where before every critic did. — justified
  5. When two critics raise the same point, the later one is ruled same as <id> and the proposer answers once. — justified
  6. Claim checks: Q: question objections, proof cells, a named non-author checker, a proof checked: line, A-M3 ids. — justified
  7. In round 1, the proposer's response table answers only the Round 0 rows. — justified

[FIRST-PRINCIPLES-REVIEWED] d5c4df9

Description read

[DESCRIPTION-READ] ae034ef8fb3372f8f513103a8f19865a1d6fe2715a66e267d52a1abcfd0d00b4

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed d5c4df9201b9218a54fee550dce7a72b4ab9d01e — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] d5c4df9

Verdict parsed from the review's SHA-scoped output markers for commit d5c4df9201b9218a54fee550dce7a72b4ab9d01e.

False positive or not applicable? A repository writer can comment:
/ai-review override fable d5c4df9201b9218a54fee550dce7a72b4ab9d01e false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Nov 9, 2026
@iamwhatever

Copy link
Copy Markdown
Contributor Author

First Principles disposition on head 1db76e0, Notes item 4 (note-copy check moved from every critic to one named critic): kept, no change.

The defect it fixes was found in the critic-roles debate as overlap-0-4: the protocol gave some duties to every critic at once (the smallest-thing entry and the note-copy check), so each critic repeated the same check and none owned it. Naming one critic on its card gives the check a single owner; the check itself and its blocking severity are unchanged. The debate closed overlap-0-4 as fixed in round 1.

Status of the other reds on this head: the BLOCK itself (product-shape change, RFC #33408 still open) is held until #33408 merges as accepted; this PR is then rebased once. Backend 3.12 shard 3 and Windows shard 3 fail on test_slim_skills_do_not_regrow_duplicate_guidance_or_wire_model_ids[kirocrew-worktree-dev-16000], a main break (that SKILL.md is 16004 bytes on main) fixed by #33795. Build Desktop failed on HTTP 504 downloads. Coverage Gate and PR Readiness follow from those.

@iamwhatever
iamwhatever force-pushed the docs/skill-critic-roles branch from 1db76e0 to 17bf3b9 Compare November 9, 2026 17:53
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 9, 2026
@iamwhatever
iamwhatever force-pushed the docs/skill-critic-roles branch from 17bf3b9 to d5c4df9 Compare November 9, 2026 18:41
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: passed Eligible automated validation passed for the current revision and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Nov 9, 2026
@KiroCrewCaptain
KiroCrewCaptain enabled auto-merge (squash) November 9, 2026 21:07

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approving. This head changes one file, src/kiro_crew/builtin_skills/goal-conductor/references/worker-debate.md (+43/-6, prose only). The review read it in full at d5c4df9. Each of the 7 hunks maps to one row of the change table in the accepted RFC (rfc-worker-debate.md §2.7). No test pins the changed wording. No blocking repo rule matches this path, and no recorded decision covers this protocol. PR Readiness is green on this head. A maintainer signed off the design through the accepted RFC #33408.

Done well: the new checker rule closes the gap the RFC debate found, where a critic could confirm its own adopted recommendation. It reuses the existing objection ids, relay step and critique step, and adds no new rounds, relays or work items.

Optional:

  • Line 157: "gives your recommended answer" reads as the Conductor's answer, because "you" means the Conductor everywhere else in this file. The RFC and line 200 both say the critic gives it. "gives its recommended answer" would fix it.
  • Line 207: the A-M3 id relies on mechanism row numbers, but the mechanism table never defines a stable M<n> row id. If rows are numbered by position, removing a row can repoint an open proof assignment. One clause saying row ids are never renumbered or reused would fix it.
  • Line 131: the new same as <id> ruling is missing from the Ruling table at lines 190 to 198.

@KiroCrewCaptain
KiroCrewCaptain merged commit 92ee944 into main Nov 9, 2026
100 of 102 checks passed
@KiroCrewCaptain
KiroCrewCaptain deleted the docs/skill-critic-roles branch November 9, 2026 21:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

readiness: passed Eligible automated validation passed for the current revision

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants