QACraft is an open-source Python library that equips developers with reusable quality assurance skills for testing AI coding agents. It provides safe adapter layers for agents such as Claude Code, GitHub Copilot, Gemini CLI, and OpenCode, along with tools for deterministic behavior evaluations. The package helps ensure reliable and consistent performance of AI-powered development tools through structured testing frameworks.
QACraft is an Agent evaluation & testing project. It focuses on evaluating and ensuring deterministic, safe behavior of AI coding agents and copilots in a reproducible way. It is built as an open-source project for developers. QACraft is open source under the MIT license. It ships for the command line and API.
Sudarshan Chaudhari builds and maintains QACraft, and it first shipped in 2026. Development happens publicly on GitHub with 48 commits in the last 90 days. Key capabilities include Safe Agent Adapters, Deterministic Evaluations, and Reusable QA Skills.
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