Prompt library
Review a diff like a senior engineer
The pageable-owner frame shifts review from style commentary to failure hunting, and the fixed severity ladder keeps nits from drowning blockers. Ordering correctness before style mirrors how strong reviewers actually allocate attention.
Last reviewed July 17, 2026
The prompt
Review the diff below as a senior engineer who will be paged when it breaks.
{{diff}}
Context: {{context}}
Review in this order and label each finding:
1. Correctness: logic errors, unhandled edge cases (empty, null, concurrent, retry), and broken invariants. Quote the exact lines.
2. Security: injection, authz gaps, secrets, unsafe deserialization, SSRF surface.
3. Data: migrations, backward compatibility, what happens to in-flight requests during deploy.
4. Tests: which changed behavior has no test, and the one test most worth adding.
5. Style: only violations of stated conventions; no taste wars.
For each finding: severity (blocker, should-fix, nit), the line reference, and a concrete fix. End with an approval verdict: approve, approve-with-nits, or request-changes, plus the single most important item.Run in idaptOpens a new chat with the prompt prefilled. Nothing sends until you press send.
Fill in the variables
| Variable | What it is | Example |
|---|---|---|
| {{diff}} | The diff or PR contents | [paste the diff] |
| {{context}} | What the change is supposed to do | adds retry with backoff to the webhook sender |