Library / Skills / SkillEvaluator
SkillEvaluator
SkillFREEMulti-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
NVIDIA/SkillEvaluator · Apache-2.0 · Python · updated today
Install with the skills CLI
612
GitHub stars
—
New stars this week
76
Forks
A · A · A
License · Quality · Upkeep
88/100
Radar score
What it helps you do
Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
Should you use it?
License · A
Permissive license. Free for personal and business use.
Quality · A
Clear README, topics, license and an active community.
Maintenance · A
Updated in the last 30 days.
Job area: Coding & DevOps. Always read the project README before you install, and test on a copy of your data first.
How to start
- 1Open the project on GitHub and read the README.
- 2Run: npx skills add NVIDIA/SkillEvaluator
- 3Ask your AI to use the new skill in plain words.
- 4Test with a small task before real work.