🛡️ Post-Deploy Watchdog

Monitors your app after deployment while your laptop is closed. Checks error rates, build status, and API health on a recurring schedule overnight.


✂️ Prompt — Copy & Paste Into Claude Code

Check the latest deployment on the main branch. Review the following:

1. Error logs — scan for any new errors, exceptions, or stack traces that appeared after the most recent deploy
2. Build status — confirm CI/CD pipeline is green and no tests are failing
3. API health — check response times and status codes for critical endpoints

If anything looks wrong:
- Summarise the issue in 2-3 sentences
- Identify the likely cause
- Suggest a fix or rollback strategy

If everything looks healthy, confirm with a short "all clear" summary.

Post the results to Slack in #deployments.

⏰ Recommended Schedule

Setting Value
Frequency Every 2 hours
Active Window 6pm → 8am (overnight after deploy)
Alternative Daily at 6am for a morning health report

💡 Why Cloud Matters Here

You deploy at 5pm and close your laptop. Local tasks can't fire. Cloud runs at 6pm, 8pm, 10pm, midnight — catching issues before your users do.