A health check endpoint answers one question fast: is this API up, and can it reach its own database. Uptime monitors, load balancers, and status pages all expect one, and building it by hand (querying the database yourself, composing the right JSON, choosing a status code) is more work than it should be for something this simple.
Building it
- Create a new GET endpoint, something like
/healthor/status. - Drag in a Health Check block. Leave Check that the database can be reached on, it runs one lightweight query, nothing heavy.
- Add a JSON Response block. Set its body to
{{health}}and its status field to{{health_status}}(or whatever variable name you gave the Health Check block, shown in its own panel). - Connect the Health Check block's two outgoing paths (marked "Ok" and "Unhealthy") both into that same Response block, since it already carries the right status either way.
That's the whole endpoint. A healthy call returns 200 with a body like {"status":"ok","checks":{"database":"ok"}}; if the database can't be reached, it returns 503 with {"status":"unhealthy","checks":{"database":"error"}} instead, exactly the status code a monitoring tool expects to treat as a real outage.
Tip: Keep a health check endpoint public (no API key required) so an external monitor can call it without needing credentials configured.