agent-evidence-guard is an open-source Python package and command-line tool for deterministically validating AI-agent completion claims. It checks claims against evidence and authority requirements, supporting safety and CI guardrails for agent developers.
agent-evidence-guard is a LLM eval & observability project. It focuses on preventing AI agents from making unsupported completion claims without required evidence or authority. agent-evidence-guard is an open-source project aimed at AI-agent developers and engineering teams. The project is open source (Apache-2.0). It ships for the command line, and it can be self-hosted.
It is developed by Ivan Varuso, and it first shipped in 2026. Development happens publicly on GitHub with 15 commits in the last 90 days. Among its 6 catalogued features are claim validation, evidence checks, and authority checks.
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