@@ -41,6 +41,28 @@ gitcrawl cluster-detail openclaw/openclaw --id <cluster-id> --member-limit 20 --
|
41 | 41 | - `invalid` |
42 | 42 | - `dirty` for PRs only |
43 | 43 | |
| 44 | +## Select small high-confidence triage candidates |
| 45 | + |
| 46 | +When asked for `X` issues or PRs to triage, `X` means qualified candidates, not sampled threads. |
| 47 | + |
| 48 | +Only list candidates that pass all gates: |
| 49 | + |
| 50 | +- small owner/surface, with a likely narrow fix and focused regression test |
| 51 | +- symptom is reproducible or provable with logs, failing test, live command, dependency contract, or current-main behavior |
| 52 | +- root cause is traceable to code with file/line and the proposed fix touches that path |
| 53 | +- no strong smell that a broader refactor, ownership rethink, migration, or product decision is the better fix |
| 54 | +- dependency-backed behavior checked against upstream docs/source/types; live or web proof used when local proof is insufficient |
| 55 | + |
| 56 | +Loop: |
| 57 | + |
| 58 | +1. Use `gitcrawl` / `gh` to gather candidate clusters. |
| 59 | +2. Read issue/PR body, comments, current code, adjacent tests, and dependency contracts. |
| 60 | +3. Try focused repro or proof. |
| 61 | +4. Reject unclear, stale, speculative, broad-refactor, or owner-ambiguous items. |
| 62 | +5. Continue until `X` qualified candidates or the bounded search is exhausted. |
| 63 | + |
| 64 | +Output only qualifying candidates, with: ref, surface, proof, cause, fix sketch, why small, expected test/gate. If none qualify, say so; do not pad. |
| 65 | + |
44 | 66 | ## Enforce the bug-fix evidence bar |
45 | 67 | |
46 | 68 | - Never merge a bug-fix PR based only on issue text, PR text, or AI rationale. |
|
