Filesystem RAG benchmarks: corpus/, train.json, evaluate_rag.py (RAGAS quality). Not for prod monitoring, latency/throughput benchmarking (use rag-perf), or evals outside this repo layout.
安装
npx skills add NVIDIA/skills --skill rag-evalnpx skills add NVIDIA/skills --skill rag-eval -a claude-code -gnpx skills add NVIDIA/skills --skill rag-eval -a codex -ggemini skills install https://github.com/NVIDIA/skills.git --path skills/rag-evalgh skill install NVIDIA/skills rag-evalgit clone --depth 1 https://github.com/NVIDIA/skills /tmp/NVIDIA__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/NVIDIA__skills/skills/rag-eval ~/.claude/skills/rag-eval # Claude Code
cp -R /tmp/NVIDIA__skills/skills/rag-eval ~/.agents/skills/rag-eval # Codex, Cursor, Gemini CLI, Copilot, OpenCode