LoopLens¶
Chrome DevTools for AI agent loops. A local-first, real-time debugger that shows your agent's execution live and warns when it repeats itself, burns tokens, retries blindly, or stops making progress.
LoopLens gives you:
- a live timeline of agent execution
- LLM-call and tool-call visibility
- retry and handoff tracking
- token and cost metrics
- loop warnings (repeated tool, no-progress, retry storm, cost spike, …)
- JSONL import/export
- a local-first UI — no login, no cloud, no API key
See it in action¶
A looping agent calls web_search over and over. As events stream in live, the
metrics climb and LoopLens flags Repeated tool call and No-progress loop
— the warning counts track the repeat count in real time, and the health score
drops to Warning.
Three ways to capture your agent¶
You don't have to rebuild your agent inside LoopLens. Pick whichever fits:
| Approach | Code in your agent | Best for |
|---|---|---|
| OpenTelemetry | none | any framework with an OTel instrumentor |
| LangGraph adapter | one callback handler | LangGraph / LangChain |
| Manual SDK | trace() / event() / @observe |
hand-rolled loops, full control |
Next steps¶
- Install LoopLens
- Quickstart — dashboard up and a demo loop in three commands
- How loop detection works