Your multi-agent pipeline failed. You don't know which agent caused it, why it made that decision, or how the failure spread. Rifft shows you.
The problem
Agent A passes bad data to Agent B. Agent B treats it as fact. By the time Agent C fails, you're three levels deep with no idea where it started. Standard logs tell you what happened. Rifft tells you why.
How it works
01
Rifft traces the failure back to its origin — not just where the pipeline crashed, but which agent's decision started the chain reaction.
02
Every agent decision is captured with its full context — the prompt state, conversation history, and tools available at the moment it chose.
03
Step through any trace, edit the agent's state at any point, and fork to test whether your fix would have worked — without paying for another full run.
Quickstart
git clone https://github.com/rifft-dev/rifft then docker compose up. Everything starts together.
pip install rifft rifft-crewai. Also works with AutoGen: rifft-autogen.
No decorators, no code changes to your agents. The adapter instruments everything on import.
localhost:3000 shows the communication graph, failure classification, and replay controls.
Open source, self-hostable, MIT licence. Running in 5 minutes.