BuzzbeeSCD/openchamber
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs(agents): weigh trusted community reviewers in triage sweeps

Browse Source 
Comments and reviews from patrick-motard and mattv8 are strong human
signals during batch triage: their approval upgrades confidence, their
concerns become findings to verify. They write free-form; sweeps map
their conclusions onto the verdict ladder.
This commit is contained in:
Bohdan Triapitsyn
2026-08-28 11:49:45 +03:00
parent 820e3194b5
commit 9d279137ce
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.agents/skills/triage-prs/SKILL.md
+2
View File
@@ -33,6 +33,8 @@ Execute the approved closes/comments with retries and ~1–2s spacing between ca
## Phase 3 — Verdict reviews ## Phase 3 — Verdict reviews
**Trusted community reviewers.** Comments and reviews from `patrick-motard` and `mattv8` are strong human signals: during any sweep, collect the PRs/issues they weighed in on, read their assessment, and carry it into the verdict — an approval from them upgrades confidence like a passing verifier; a concern from them is a finding to verify, never to ignore. They write free-form; map their conclusion onto the verdict ladder rather than expecting the format.
The review bot's `review:*` labels are a pre-sort, not a verdict: `review:ready` PRs go first (the bot found no code defects — likely MERGE/MERGE-THEN-FIX), `review:blocked` ones carry a bot comment whose findings the verdict review verifies rather than rediscovers. Bot labels never replace the pr-review pass — the bot cannot judge product fit or maintainability scope. The review bot's `review:*` labels are a pre-sort, not a verdict: `review:ready` PRs go first (the bot found no code defects — likely MERGE/MERGE-THEN-FIX), `review:blocked` ones carry a bot comment whose findings the verdict review verifies rather than rediscovers. Bot labels never replace the pr-review pass — the bot cannot judge product fit or maintainability scope.
Split the clean pool smallest-first (tiny diffs are fast wins and most likely mergeable). Fan out subagents in batches of ~10 PRs each; every subagent receives the full `pr-review` skill text as its instructions plus its PR numbers, reads real diffs (`gh pr view`, `gh pr diff`) and the local checkout, and returns per-PR verdict blocks in the skill's output format. Split the clean pool smallest-first (tiny diffs are fast wins and most likely mergeable). Fan out subagents in batches of ~10 PRs each; every subagent receives the full `pr-review` skill text as its instructions plus its PR numbers, reads real diffs (`gh pr view`, `gh pr diff`) and the local checkout, and returns per-PR verdict blocks in the skill's output format.