Skip to content
Observability

Hire an agent

Run a read-only monitoring routine in your own agent harness.

This guide explains how to run a Supabase monitoring agent in your own harness. Each agent is a prompt plus a schedule. It reads project data and reports findings. It does not change the project.

Choose a routine#

Start with one monitor. Add another only when the project needs a different source or cadence.

MonitorWhat it watchesDefault cadenceUse it when
Health monitorAPI and Auth server errors, error-rate spikes, connection pressureHourlyYou need incident detection and regular feedback loops
Security monitorSecurity Advisor findings, authentication and authorization failuresDailyYou need a regular access-control and configuration review
Performance monitorSlow queries, lock waits, long-running sessions, Performance Advisor findingsHourlyYou need query and database performance checks
Capacity monitorRequest, error, storage, table, and connection growthDailyYou need to identify growth before it reaches a resource limit

For a small project, run the most relevant routine daily or weekly and include the other categories in its prompt. Split it into specialized monitors only when you need different owners, schedules, or alert thresholds.

Run the routine#

  1. Connect the Supabase MCP server with project_ref and read_only=true.
  2. Open the monitor that matches the job from the table above.
  3. Set it up in Claude, Codex, Cursor, or copy the prompt into another harness.
  4. Run it on demand first. Then put the same prompt on a schedule.

Each agent page describes what it will output. Send those findings through the connections your harness already has, such as Linear in Codex.

Scheduled tasks start a fresh context on every run, so the prompt is self-contained. Review the first runs before you rely on the schedule.