| name | prod-readiness-review |
| description | Check production readiness (rollback, monitoring, SLOs, runbooks, load tests, config sanity, migrations) for React Native + Next.js. Use to gate releases and reduce incident risk. |
| metadata | {"short-description":"Production readiness review"} |
Production Readiness Review
Use this skill to assess release readiness for web and mobile apps with a focus on operational safety.
Key behavior:
- Prefer repo evidence over generic advice (existing runbooks, CI scripts, infra code).
- When evidence is missing, mark items as
UNKNOWN rather than assuming.
- Produce a crisp “go/no-go” summary: blockers vs risks vs nice-to-haves.
Workflow
- Identify release artifacts and deployment paths.
- Next.js: build outputs, hosting target, environment config.
- React Native: build pipelines, store release steps, OTA updates.
- Verify operational readiness.
- Monitoring: logs, metrics, traces, crash reporting (web + mobile).
- Alerting: on-call notifications and severity thresholds.
- Runbooks: incident response and rollback steps.
- SLOs: performance or availability targets.
- Validate data and config safety.
- Environment variables and secrets are managed securely.
- Feature flags for risky changes.
- Migration safety: backward compatibility and rollback plan.
- Config sanity: required env vars documented, safe defaults, non-prod vs prod behavior is explicit.
- Check release quality gates.
- CI is green and required tests are present.
- Smoke tests or canary plans exist.
- Versioning and changelog practices are followed.
- For mobile: build numbers/versions incremented; release notes; staged rollout plan if supported.
- For web: cache invalidation strategy (CDN/ISR), and safe deploy order if multiple services.
- Confirm "break-glass" and recovery.
- Rollback is documented and practical (not theoretical).
- If migrations exist, confirm a rollback or forward-fix plan exists.
- Verify incident owner/on-call path exists (even if lightweight).
Readiness Checklist Template
Use this structure and fill it with PASS/FAIL/UNKNOWN and evidence:
- Deploy:
- Web deploy mechanism: PASS/FAIL/UNKNOWN (evidence: file/doc)
- Mobile release mechanism: PASS/FAIL/UNKNOWN (evidence: file/doc)
- Observability:
- Web errors + traces: PASS/FAIL/UNKNOWN (evidence)
- Mobile crash reporting: PASS/FAIL/UNKNOWN (evidence)
- Alerts routed to humans: PASS/FAIL/UNKNOWN (evidence)
- Safety:
- Rollback plan: PASS/FAIL/UNKNOWN (evidence)
- Feature flag for risky behavior: PASS/FAIL/UNKNOWN (evidence)
- Migration plan (backward compatible): PASS/FAIL/UNKNOWN (evidence)
- Quality gates:
- CI required checks: PASS/FAIL/UNKNOWN (evidence)
- Smoke/canary: PASS/FAIL/UNKNOWN (evidence)
- Secrets management: PASS/FAIL/UNKNOWN (evidence)
Output Format
Provide:
- Readiness Checklist (pass/fail/unknown)
- Blockers (must fix before release)
- Risks (acceptable with mitigation)
- Recommended Next Steps
Notes
- Do not invent a release process; use repo evidence.
- If key artifacts are missing, call them out explicitly.