agent-polygraph is a command-line tool that evaluates AI agent completions for internal contradictions, helping developers and researchers detect hallucinations and improve agent reliability. It is framework-agnostic and integrates with observability tools. Distributed under the MIT license.
agent-polygraph sits in PulseGate's Agent evaluation & testing category. It focuses on detecting and flagging contradictions or hallucinations in AI agent outputs to improve reliability. It is built as an open-source project for AI developers and researchers. The project is open source (MIT). It runs on the command line.
It is developed by najemwehbe, and it first shipped in 2026. The project is developed in the open on GitHub with 2 commits in the last 90 days. Among its 5 catalogued features are lie detection, agent evaluation, and observability.
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