altrace-dev-role/rashomon

13 stars · Last commit 2026-10-04

Verify what your Claude Code agent actually did. Rashomon keeps an independent record of every tool call, subagent, failed command, and flags when the agent's summary says all tests pass but something failed. Open-source AI agent verification, hallucination detection, and audit tool. Local-only, no telemetry.

README preview

# rashomon

**Your coding agent writes its own account of what it did. `rashomon` writes a second one, independently of the agent's narration, and puts the two side by side.**

**[Join the Rashomon Discord](https://discord.gg/Qg3hybAHpU)**

*For Claude Code only, on macOS and Linux. Alpha: v1.1.0.*

![Claude says all tests pass; rashomon's end-of-turn line reports a recorded failure, and an excerpt of the report shows it](docs/images/rashomon-example.png)

## In plain words

When Claude Code finishes a task, it tells you what it did: "Done, all tests
pass." That summary comes from the agent itself.

rashomon keeps its own record of every step the agent takes, like each command
it runs and each file it edits, and whether each one worked. It puts that record
beside the agent's summary: if a step failed and the closing message uses none
of rashomon's failure words, it tells you at the end of the turn. Every report
also lists what subagents did, which the main conversation does not show.

View full repository on GitHub →