mistral-prod-checklist
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".
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 2.0k
- Forks
- 233
01 前言 Front Matter
- name
- mistral-prod-checklist
- author
- Jeremy Longshore <jeremy@intentsolutions.io>
- license
- MIT
- version
- 1.0.0
- description
- 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".
- allowed-tools
- Read, Bash(kubectl:*), Bash(curl:*), Grep
03 SKILL.md 正文
Mistral AI Production Checklist
Overview
Complete checklist for deploying Mistral AI integrations to production.
Prerequisites
- Staging environment tested and verified
- Production API keys available
- Deployment pipeline configured
- Monitoring and alerting ready
Instructions
Step 1: Pre-Deployment Configuration
Secrets & Configuration
- [ ] Production API key in secure vault (AWS Secrets Manager, GCP Secret Manager, etc.)
- [ ] Environment variables set in deployment platform
- [ ] API key validated with test request
- [ ] Fallback configuration ready (if Mistral unavailable)
Verify API Key:
# Test production key
curl -H "Authorization: Bearer ${MISTRAL_API_KEY_PROD}" \
https://api.mistral.ai/v1/models | jq '.data[].id'
Step 2: Code Quality Verification
Code Review
- [ ] All tests passing (
npm test) - [ ] No hardcoded credentials (
grep -r "MISTRAL_API_KEY" src/) - [ ] Error handling covers all error types (401, 429, 500, etc.)
- [ ] Rate limiting/backoff implemented
- [ ] Logging is production-appropriate (no sensitive data)
- [ ] TypeScript types are strict
Run Verification:
# Full test suite
npm test
# Type checking
npm run typecheck
# Check for hardcoded secrets
grep -r "sk-" src/ --include="*.ts" --include="*.js"
# Lint
npm run lint
Step 3: Infrastructure Setup
Health Checks
- [ ] Health check endpoint includes Mistral connectivity test
- [ ] Readiness probe configured
- [ ] Liveness probe configured
// /health endpoint
app.get('/health', async (req, res) => {
const mistralHealth = await checkMistralHealth();
res.status(mistralHealth.healthy ? 200 : 503).json({
status: mistralHealth.healthy ? 'healthy' : 'degraded',
services: {
mistral: mistralHealth,
},
timestamp: new Date().toISOString(),
});
});
async function checkMistralHealth() {
const start = Date.now();
try {
await client.models.list();
return { healthy: true, latencyMs: Date.now() - start };
} catch (error) {
return { healthy: false, latencyMs: Date.now() - start, error: 'API unreachable' };
}
}
Monitoring & Alerting
- [ ] Prometheus/Datadog metrics configured
- [ ] Alert rules defined (error rate, latency, rate limits)
- [ ] Dashboard created
- [ ] On-call rotation scheduled
Step 4: Resilience Patterns
Circuit Breaker
- [ ] Circuit breaker pattern implemented
- [ ] Graceful degradation configured
- [ ] Fallback responses defined
class MistralCircuitBreaker {
private failures = 0;
private lastFailure?: Date;
private isOpen = false;
private readonly failureThreshold = 5;
private readonly resetTimeout = 60000; // 1 minute
async call<T>(fn: () => Promise<T>, fallback?: () => T): Promise<T> {
// Check if circuit is open
if (this.isOpen) {
if (Date.now() - (this.lastFailure?.getTime() || 0) > this.resetTimeout) {
this.isOpen = false;
this.failures = 0;
} else {
if (fallback) return fallback();
throw new Error('Mistral service temporarily unavailable');
}
}
try {
const result = await fn();
this.failures = 0;
return result;
} catch (error: any) {
this.failures++;
this.lastFailure = new Date();
if (this.failures >= this.failureThreshold) {
this.isOpen = true;
console.error('Mistral circuit breaker opened');
}
if (fallback && this.isOpen) return fallback();
throw error;
}
}
}
Step 5: Documentation Requirements
- [ ] Incident runbook created
- [ ] Key rotation procedure documented
- [ ] Rollback procedure documented
- [ ] On-call escalation path defined
- [ ] API usage limits documented
Step 6: Deploy with Gradual Rollout
# Pre-flight checks
echo "=== Pre-flight Checks ==="
# 1. Check Mistral status
echo -n "Mistral API: "
curl -s -o /dev/null -w "%{http_code}" \
-H "Authorization: Bearer ${MISTRAL_API_KEY_PROD}" \
https://api.mistral.ai/v1/models
echo ""
# 2. Check staging health
echo -n "Staging health: "
curl -sf https://staging.yourapp.com/health | jq -r '.status'
# 3. Verify tests pass
npm test
# Gradual rollout
echo "=== Starting Gradual Rollout ==="
# Deploy to canary (10% traffic)
kubectl set image deployment/mistral-app app=image:new --record
kubectl rollout pause deployment/mistral-app
# Monitor for 10 minutes
echo "Monitoring canary for 10 minutes..."
sleep 600
# Check error rates
kubectl logs -l app=mistral-app --since=10m | grep -c "error" || echo "0 errors"
# If healthy, continue to 50%
kubectl rollout resume deployment/mistral-app
kubectl rollout pause deployment/mistral-app
sleep 300
# Complete rollout to 100%
kubectl rollout resume deployment/mistral-app
kubectl rollout status deployment/mistral-app
Step 7: Post-Deployment Verification
# Verify deployment
echo "=== Post-Deployment Verification ==="
# Health check
curl -sf https://yourapp.com/health | jq
# Test chat endpoint
curl -X POST https://yourapp.com/api/chat \
-H "Content-Type: application/json" \
-d '{"messages": [{"role": "user", "content": "Hello"}]}' | jq
# Check logs for errors
kubectl logs -l app=mistral-app --since=5m | grep -i error
Output
- Deployed Mistral AI integration
- Health checks passing
- Monitoring active
- Rollback procedure documented
Alert Configuration
| 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 |
Rollback Procedure
# Immediate rollback
echo "=== Emergency Rollback ==="
# 1. Rollback deployment
kubectl rollout undo deployment/mistral-app
# 2. Verify rollback
kubectl rollout status deployment/mistral-app
# 3. Check health
curl -sf https://yourapp.com/health | jq
# 4. Notify team
# Send alert to Slack/PagerDuty
Resources
Next Steps
For version upgrades, see mistral-upgrade-migration.
04 同仓库更多 Skills
theme-factory
Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.
web-artifacts-builder
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
neuropixels-analysis
Neuropixels neural recording analysis. Load SpikeGLX/OpenEphys data, preprocess, motion correction, Kilosort4 spike sorting, quality metrics, Allen/IBL curation, AI-assisted visual analysis, for Neuropixels 1.0/2.0 extracellular electrophysiology. Use when working with neural recordings, spike sorting, extracellular electrophysiology, or when the user mentions Neuropixels, SpikeGLX, Open Ephys, Kilosort, quality metrics, or unit curation.
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
notebooklm
Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.