Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute CodeRabbit production readiness checklist for org-wide deployment. Use when preparing to enforce CodeRabbit reviews, going live with required checks, or auditing CodeRabbit configuration before making it a merge gate. Trigger with phrases like "coderabbit production", "coderabbit go-live", "coderabbit launch checklist", "coderabbit readiness", "coderabbit pre-launch".
.claude/skills/jeremylongshore-coderabbit-prod-checklist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 78% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 121% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 48% | 0% |
Turn a pilot into an explicit production decision. Each gate needs an owner, evidence, failure criterion, and rollback.
references/official-docs.md and re-check any time-sensitive contract before execution.Treat Git-provider sessions, CodeRabbit web sessions, CLI credentials, and CodeRabbit API keys as separate credentials. Use only an already-approved session or secret-manager reference, never print a secret, and do not place credentials in .coderabbit.yaml, source files, logs, or deliverables.
Named repository, security, privacy, billing, and operations owners approve their own gates. Keep analysis and drafts local until approval is explicit, and record who approved the action and its scope.
A readiness matrix, receipts, risks, continuity plan, rollback, and decision. Include source dates, unknowns, and the exact boundary between observed fact and recommendation.
| Condition | Response | |---|---| | Current contract is unclear or docs disagree | Stop mutation, cite both sources, and request owner resolution. | | Required access or approval is missing | Produce a draft and evidence plan only. | | Validation or pilot behavior differs from expectation | Restore the prior state and retain the failed evidence. | | Output contains secrets or private code | Stop, quarantine the artifact, redact it, and notify the data owner. |
Issue conditional go-live limited to pilots.
Reject enforcement when manual continuity is untested.
references/official-docs.md.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 16,320 | 16,143 | -1% | 1 | 1 | 0% | 3,270 | 5,512 | +69% | 0 | 0 | — |
case-02 | fail→pass | 29,462 | 12,477 | -58% | 1 | 1 | 0% | 5,104 | 4,648 | -9% | 0 | 0 | — |
case-03 | pass→pass | 4,735 | 4,003 | -15% | 1 | 1 | 0% | 952 | 3,113 | +227% | 0 | 0 | — |
case-04 | pass→pass | 4,831 | 4,758 | -2% | 1 | 1 | 0% | 888 | 3,105 | +250% | 0 | 0 | — |
case-05 | pass→pass | 6,901 | 3,006 | -56% | 1 | 1 | 0% | 1,161 | 2,781 | +140% | 0 | 0 | — |
case-06 | pass→pass | 5,996 | 2,903 | -52% | 1 | 1 | 0% | 936 | 2,728 | +191% | 0 | 0 | — |
case-07 | pass→pass | 2,463 | 2,742 | +11% | 1 | 1 | 0% | 444 | 2,659 | +499% | 0 | 0 | — |
case-08 | pass→pass | 10,693 | 8,766 | -18% | 1 | 1 | 0% | 1,818 | 3,782 | +108% | 0 | 0 | — |
case-09 | pass→pass | 11,075 | 3,882 | -65% | 1 | 1 | 0% | 1,924 | 2,859 | +49% | 0 | 0 | — |
case-10 | pass→pass | 4,733 | 2,569 | -46% | 1 | 1 | 0% | 747 | 2,665 | +257% | 0 | 0 | — |
case-11 | pass→pass | 11,155 | 5,950 | -47% | 1 | 1 | 0% | 1,804 | 3,229 | +79% | 0 | 0 | — |
case-12 | fail→pass | 13,012 | 12,980 | -0% | 1 | 1 | 0% | 2,756 | 4,914 | +78% | 0 | 0 | — |
case-13 | pass→pass | 3,120 | 2,204 | -29% | 1 | 1 | 0% | 569 | 2,538 | +346% | 0 | 0 | — |
case-14 | pass→pass | 5,467 | 4,897 | -10% | 1 | 1 | 0% | 1,100 | 3,112 | +183% | 0 | 0 | — |
case-15 | fail→fail | 10,276 | 6,978 | -32% | 1 | 1 | 0% | 2,296 | 3,605 | +57% | 0 | 0 | — |
case-16 | pass→pass | 8,008 | 2,217 | -72% | 1 | 1 | 0% | 1,405 | 2,532 | +80% | 0 | 0 | — |
case-17 | fail→pass | 7,479 | 1,541 | -79% | 1 | 1 | 0% | 1,072 | 2,368 | +121% | 0 | 0 | — |
case-18 | fail→pass | 12,355 | 4,116 | -67% | 1 | 1 | 0% | 1,963 | 2,898 | +48% | 0 | 0 | — |
case-19 | pass→pass | 9,188 | 5,596 | -39% | 1 | 1 | 0% | 1,575 | 3,178 | +102% | 0 | 0 | — |
case-20 | pass→pass | 6,993 | 5,370 | -23% | 1 | 1 | 0% | 1,248 | 3,135 | +151% | 0 | 0 | — |
case-21 | pass→pass | 6,309 | 2,652 | -58% | 1 | 1 | 0% | 1,208 | 2,689 | +123% | 0 | 0 | — |
case-22 | pass→pass | 1,913 | 1,154 | -40% | 1 | 1 | 0% | 326 | 2,399 | +636% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.