agentfork
github.com/bimal1023/agentfork- Category
- Developer Tools
- Rank
- No. 2406Tools index
- Pricing
- Open Source
- Type
- TOOL
- Date
About
A Python tool that records every LLM and tool call of an agent, replays earlier steps from cache, and runs many variants of one step several times each. Variants can change the prompt, model or tool results. This gives pass rates instead of single-run anecdotes and avoids paying for steps that did not change. It runs locally with no account or server.
Tags
llmai-agentsdebuggingevaluationreplaypython
Tech Stack
Python
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.