codeintegrity-ai/mutahunter

Leverages LLM models (GPT-4o, etc.) to intelligently generate code mutants across any programming language, executing them against existing test suites to measure test effectiveness. Generates mutants by modifying source code semantically, then runs arbitrary test commands to detect which mutants survive, providing mutation coverage metrics and cost tracking. Supports polyglot workflows through configurable test commands and source/test file paths, enabling integration with Maven, npm, pytest, and other language-specific test frameworks.

288 stars and 53 monthly downloads. No commits in the last 6 months. Available on PyPI.