Observe the data

This guide lists the project data you can query. Each source page lists where to read that source. To pick up a signal from this data, see Detecting.

Logs#

Request, database, Auth, Storage, Realtime, and function events in ClickHouse.

Query them with SQL in Query and filter logs from the Logs Explorer, MCP query_logs, or the Management API. Filter events without SQL in Logs. See the Logs field reference for sources and fields.

The CLI does not query ClickHouse logs. Call the Management API from a script, or inspect the database for Postgres diagnostics.

Metrics #

Prometheus-compatible CPU, IO, WAL, connections, and query stats. Scrape the Metrics API for custom dashboards, alerting, or retention beyond Studio. Chart a subset of the same window in Reports.

Database#

Live Postgres statistics such as bloat, cache hit rate, blocking sessions, index usage, and slow queries. Run the same checks from the SQL Editor, MCP execute_sql, or supabase inspect db. See Inspect the database.

Advisors#

Deterministic security and performance findings. Pull them from Studio, MCP get_advisors, supabase db advisors, or the Management API. See Advisors.

Reports#

Studio dashboards for API, Auth, Storage, Realtime, and database signals. Use them to pick a time window or resource, then follow Detecting. See Reports.