auditable is an Apache-2.0 Python package that records the evidence and state behind AI-agent decisions. It supports replaying decisions against live state and rolling back committed actions when they no longer remain valid, for developers building reliable agent systems.
auditable is an Agent monitoring & governance project. Tracking, replaying, and safely rolling back AI-agent decisions when their underlying conditions change. auditable is an open-source project aimed at AI-agent developers and platform engineers. The project is open source (Apache-2.0). It ships for the command line and API, and it can be self-hosted.
It is developed by Yuan Zhao, and it first shipped in 2026. Development happens publicly on GitHub with 21 stars and 28 commits in the last 90 days. Key capabilities include decision recording, evidence capture, and decision replay. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do