Show HN: AgentLens – Open-source observability for AI agents
2 コメント
The IAM framing is a good starting point but I think it undersells the core difficulty: IAM assumes the identity of the actor is stable and legible. With agents, the "actor" at step 15 of a multi-step run may have meaningfully different context and effective goals than at step 1 — same session, different behavior.
Policy enforcement at runtime has to grapple with this. A rule like "don't write to production DBs" is easy. A rule like "don't proceed if your current action contradicts the intent you stated 3 steps ago" requires the agent to have exposed its intent in a queryable form throughout the run — which most don't.
Auditability is the easier problem. Runtime intervention based on goal drift is where I think the field is genuinely unsolved.
The topology graph and time-travel replay are genuinely useful additions — most tools in this space flatten agent runs into a linear trace, losing the branching structure entirely.
One thing I'd be curious about: how does AgentLens handle the difference between an agent that failed to execute vs one that executed but drifted from its stated goal? Execution traces capture the former well. The latter requires knowing what the agent intended before the tool call happened — which means you need to log the pre-action intent state, not just the post-action result.
This matters most for debugging silent failures — cases where all tool calls succeed and costs look normal, but the agent quietly pursued the wrong sub-goal.
I built AgentLens because debugging multi-agent systems is painful. LangSmith is cloud-only and paid. Langfuse tracks LLM calls but doesn't understand agent topology — tool calls, handoffs, decision trees.
AgentLens is a self-hosted observability platform built specifically for AI agents:
- *Topology graph* — see your agent's tool calls, LLM calls, and sub-agent spawns as an interactive DAG - *Time-travel replay* — step through an agent run frame-by-frame with a scrubber timeline - *Trace comparison* — side-by-side diff of two runs with color-coded span matching - *Cost tracking* — 27 models priced (GPT-4.1, Claude 4, Gemini 2.0, etc.) - *Live streaming* — watch spans appear in real-time via SSE - *Alerting* — anomaly detection for cost spikes, error rates, latency - *OTel ingestion* — accepts OTLP HTTP JSON, so any OTel-instrumented app works
Works with LangChain, CrewAI, AutoGen, LlamaIndex, and Google ADK.
Tech: React 19 + FastAPI + SQLite/PostgreSQL. MIT licensed. 231 tests, 100% coverage.
Demo GIF and screenshots in the README.
GitHub: https://github.com/tranhoangtu-it/agentlens-observe Docs: https://agentlens-observe.pages.dev
I'd love feedback on the trace visualization approach and what features matter most for your agent debugging workflow.