mgechev/skillgrade
A testing framework that runs unit-test-style evaluations of AI agent skills against configurable rubrics and task definitions.
Not currently ranked — collecting fresh signals.
star history
Skillgrade provides a framework for evaluating AI agent capabilities by executing tasks defined in eval.yaml and grading agent responses. It supports multiple agent backends (Claude, Codex, Gemini, OpenCode) and two grader types: deterministic (exact match) and llm_rubric (LLM-based evaluation). Users scaffold eval configs with AI assistance or manually, run configurable trial counts (smoke/reliable/regression), and preview results via CLI or browser UI.
Frequently asked
- What is mgechev/skillgrade?
- A testing framework that runs unit-test-style evaluations of AI agent skills against configurable rubrics and task definitions.
- Is skillgrade open source?
- Yes — mgechev/skillgrade is open source, released under the MIT license.
- What language is skillgrade written in?
- mgechev/skillgrade is primarily written in TypeScript.
- How popular is skillgrade?
- mgechev/skillgrade has 637 stars on GitHub.
- Where can I find skillgrade?
- mgechev/skillgrade is on GitHub at https://github.com/mgechev/skillgrade.