LAB · Agents
Agent Trace Debugger
Analyze a JSON trace for cycles, duplicate calls, missing timeouts, unhandled errors, and privilege escalation.
- Suggested time
- 15 min
- Output artifact
- Root-cause hypotheses and concrete missing controls.
- Privacy
- The trace never leaves the browser. Remove secrets even when working locally.
Local educational tool. Validate every result before production use.
When to use it
Trace debugger
Use it when an agent repeats actions, stalls, ignores errors, or consumes more steps than planned.
How to use it well
- 1
Enter a real case without secrets or personal data.
- 2
Run, inspect, and challenge the assumptions; the output is not automatic truth.
- 3
Save the artifact in your project and turn it into a reviewable test or decision.
Related foundations
Loop Engineering
01 — State, budgets, and termination
The minimum loop observes state, decides, acts, and observes again. The professional version adds a transition contract, verifiable budgets, and a terminal function that does not depend on optimistic model text.
Open foundation3 min02 — Errors, idempotency, and resumption
The hardest failures happen between systems: the provider accepts an action, the network drops before the response, and the loop does not know whether to repeat it. No prompt resolves that ambiguity; identity, receipts,
Open foundation5 min08 — Agent Evaluation: Outcome, Trajectory, Cost, and Safety
This topic defines review signals for the mini-project and prepares the field project evaluation.
Open foundation