DebuggerFind exactly where
Find exactly where
a request broke.
Drop in a trace ID or span ID and step through every hop the request made — the exact service, query, or call where things went wrong, not just that something did.
~/debugger
trace_id:8f3a1e2c-9b4d-4a11-8e2f-1c7d0a55e901
POST /v1/agents/run
1,840msplanQuery (groq/llama-3.1-8b)
142msquery_clickhouse: traces_hourly
380msquery_database: readonlyPool
1,240msSET LOCAL app.current_org_id
1,180msRoot cause:
SET LOCAL app.current_org_id executed outside its transaction — RLS policy blocked the read for 1.18s before failing.Why it matters
Stop guessing which service dropped the ball
Trace & span ID lookup
Paste a trace ID from a log line, a support ticket, or an alert and land directly on that exact request.
Full request timeline
Every span the request touched, in order, with timing — so a slow or failed hop is visible immediately, not inferred.
Root-cause isolation
The debugger flags the span that actually failed or stalled, instead of leaving you to scroll through a wall of logs.