📈
Database Health Watch CrewAI
blueprint · 3 agents · 2 MCP servers
What it does
Runs a set of health queries against the database on a schedule, compares against normal ranges, and when something moves it explains what changed and posts to Slack with enough context that whoever's on call knows if it matters. Signal, not just an alert.
Agent flow
1 · Poller
Run the health queries
postgres: query
2 · Detector
Compare to normal ranges
no tools · reasoning
3 · Notifier
Explain and post to Slack
slack: post_message
fork & run
# fork into your project, wire your own servers
npx socketcat fork db-health-watch