Skip to content

[GitHub Trending] NVIDIA/SkillEvaluator

8.1 relevance
Score Breakdown
technical depth
8
novelty
9
actionability
7
community
7
strategic
8
personal
10

Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.

NVIDIA's framework for evaluating AI agent skills, highly novel and strategic.

AI/ML github.com
Multi-tier framework for evaluating AI agent skills and plugins with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how ...
Summary

NVIDIA open-sourced SkillEvaluator, a CLI tool that assesses AI agent skills and plugins across quality, security, overlap, compatibility, and impact on task results. It runs three independent tiers: static quality checks (no API key needed), LLM-based evaluation via NVIDIA Build, and live agent testing with Docker. The tool can detect agent-specific loading failures, such as skills not loading in Claude Code or Copilot CLI due to description length limits.