claude-bot/EJClaw
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Tighten reviewer prompt output scope

Browse Source
This commit is contained in:
ejclaw
2026-05-17 09:05:34 +09:00
parent 82c0b1bc85
commit 46b803170d
2 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
prompts/claude-paired-room.md
View File

@@ -42,5 +42,7 @@ If you see a materially better design, debugging path, or scoping choice, propos
- Stagnation: **Spinning** (same error 3+), **Oscillation** (alternating approaches), **Diminishing returns** (shrinking improvement), **No progress** (discussion without change) — name the pattern and report: **Status**, **Attempted**, **Recommendation** - Stagnation: **Spinning** (same error 3+), **Oscillation** (alternating approaches), **Diminishing returns** (shrinking improvement), **No progress** (discussion without change) — name the pattern and report: **Status**, **Attempted**, **Recommendation**
- Implementation, commits, and pushes require agreement from both sides. Either can veto - Implementation, commits, and pushes require agreement from both sides. Either can veto
- Keep reviews concise — approve quickly when there is nothing to critique, and keep alternative proposals short and actionable - Keep reviews concise — approve quickly when there is nothing to critique, and keep alternative proposals short and actionable
- Keep reviewer output owner-facing. Do not draft user-facing messages, user instructions, or Discord mentions; tell the owner what to do or approve the owner to finalize
- On approval, prefer 3-6 lines: status, blocking findings if any, key evidence, and the next owner action. Do not explain background theory unless it changes the decision
- Do not carry over old ledgers. In reviewer finals, include only blockers, evidence, and follow-ups that directly affect the current task. Omit stale "remaining items", observations, potential follow-ups, deployment backlogs, and prior-task status tables unless the user explicitly asks for them again - Do not carry over old ledgers. In reviewer finals, include only blockers, evidence, and follow-ups that directly affect the current task. Omit stale "remaining items", observations, potential follow-ups, deployment backlogs, and prior-task status tables unless the user explicitly asks for them again
- Never mention or tag the user (@username) during the owner↔reviewer loop — the system handles escalation automatically. User is only notified when all resolution paths (including arbiter) are exhausted - Never mention or tag the user (@username) during the owner↔reviewer loop — the system handles escalation automatically. User is only notified when all resolution paths (including arbiter) are exhausted

2
src/platform-prompts.test.ts
View File

@@ -83,6 +83,8 @@ describe('platform-prompts', () => {
expect(codexPairedPrompt).toContain( expect(codexPairedPrompt).toContain(
'Separate correctness issues from improvement ideas', 'Separate correctness issues from improvement ideas',
); );
expect(codexPairedPrompt).toContain('Keep reviewer output owner-facing');
expect(codexPairedPrompt).toContain('prefer 3-6 lines');
expect(codexPairedPrompt).not.toContain('owner-side paired agent'); expect(codexPairedPrompt).not.toContain('owner-side paired agent');
const failoverPlatformPrompt = fs.readFileSync( const failoverPlatformPrompt = fs.readFileSync(