AgentOps-AI/agentops
A Python library that records each agent run, including LLM calls, tool calls and cost per session, and draws it as a replayable graph. It needs a few decorators or one init line and hooks into CrewAI, LangGraph, LangChain and LlamaIndex.
Not tested yetTested by others
How we know
Someone outside the project tested it. Their result is below, with a link to the original post. We have not tested it ourselves.
Good for
Python developers running agents who want to see where a run went wrong and what it cost.
Not for
Non-Python stacks. The hosted dashboard sends your traces to the vendor unless you self-host it.
Sources
12% added latency over an uninstrumented baseline (Langfuse 15%, Laminar 5%, LangSmith about 0%)
AIMultiple (Cem Dilmegani) · · how: 100 identical queries through a 5-stage LangChain multi-agent travel planner, each tool instrumented one at a time
https://aimultiple.com/agentic-monitoring
Repo
- Stars
- None
- License
- MITUse it, change it, sell it, closed-source included. Keep the author's copyright notice.
- Maturity
- active
- Last push
- Last checked
- 2026-11-13