Execute Mistral AI production deployment checklist and rollback procedures. Use when deploying Mistral AI integrations to production, preparing for launch, or implementing go-live procedures. Trigger with phrases like "mistral production", "deploy mistral", "mistral go-live", "mistral launch checklist".
From mistral-packnpx claudepluginhub nickloveinvesting/nick-love-plugins --plugin mistral-packThis skill is limited to using the following tools:
references/implementation.mdGuides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Details PluginEval's skill quality evaluation: 3 layers (static, LLM judge), 10 dimensions, rubrics, formulas, anti-patterns, badges. Use to interpret scores, improve triggering, calibrate thresholds.
Complete checklist for deploying Mistral AI integrations to production. Covers credential verification, code quality checks, health endpoints, circuit breaker resilience, gradual rollout, and rollback procedures.
Store production API key in secure vault (AWS Secrets Manager, GCP Secret Manager). Set environment variables in deployment platform. Validate key with test request to https://api.mistral.ai/v1/models. Prepare fallback configuration.
Run full test suite, typecheck, and lint. Scan for hardcoded credentials (grep -r "sk-" src/). Verify error handling covers 401, 429, and 500+ status codes. Confirm rate limiting/backoff is implemented and logging excludes sensitive data.
Implement /health endpoint that tests Mistral API connectivity with latency measurement. Configure readiness and liveness probes. Set up Prometheus/Datadog metrics, alert rules (error rate, latency, rate limits), and dashboard.
Build MistralCircuitBreaker with failure threshold (5), reset timeout (60s), and fallback support. Circuit opens after threshold failures and auto-resets after timeout. Provides graceful degradation when Mistral is unavailable.
Create incident runbook, key rotation procedure, rollback procedure, on-call escalation path, and API usage limits documentation.
Deploy to canary (10% traffic), monitor 10 minutes for errors, expand to 50%, then complete to 100%. Use kubectl rollout pause/resume for staged deployment.
Hit health endpoint, test chat endpoint with sample request, check logs for errors. Verify monitoring dashboards show expected metrics.
| Alert | Condition | Severity |
|---|---|---|
| API Down | 5xx errors > 10/min | P1 - Critical |
| High Latency | p99 > 5000ms for 5min | P2 - High |
| Rate Limited | 429 errors > 5/min | P2 - High |
| Auth Failures | 401 errors > 0 | P1 - Critical |
| Circuit Open | Circuit breaker triggered | P2 - High |
| Issue | Detection | Resolution |
|---|---|---|
| Deployment failure | kubectl rollout status | Rollback with kubectl rollout undo |
| Health check failing | 503 response | Check Mistral API status, verify credentials |
| Circuit breaker open | Alert triggered | Investigate Mistral availability, wait for reset |
| High error rate | Monitoring alert | Check logs, consider rollback |
set -euo pipefail
curl -sf https://yourapp.com/health | jq '.services.mistral'
set -euo pipefail
kubectl rollout undo deployment/mistral-app
kubectl rollout status deployment/mistral-app
curl -sf https://yourapp.com/health | jq
See detailed implementation for advanced patterns.