NVIDIA open-sources SkillEvaluator, a multi-tier framework for evaluating AI agent skills
NVIDIA · github · 2026-11-15
NVIDIA released SkillEvaluator on GitHub, a Python multi-tier framework for evaluating AI agent skills that quickly reached 1,333 stars. It provides quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation measuring how skills actually affect agent behavior. Topics cover agent security, Claude Code, Codex, plugin evaluation and benchmarks — a systematic eval and security-scanning layer for the agent skills ecosystem.
More from coding & agent
- Cutting a Research Agent From 12 Tools to 3 Boosted Success Rate From 52% to 84% — Such_Reflection8996 · 2026-11-15
- Bypassing Opus 5.5 refusals with a 'boss' agent: multi-agent setups revive the confused deputy flaw — rickasaurus · 2026-11-15
- Claude-BugHunter: Claude Code skill bundle with 82 skills for bug hunting hits 5.9k stars — elementalsouls · 2026-11-15
- Anthropic says start most coding work on Opus 5.5, not Fable 5.2 — effort beats model size — blaizedsouza · 2026-11-15
- Three shifts since spring: one heavy user's 'How I use AI in November 2026' — TheTuringPost · 2026-11-15
- Every team builds a video editor and writing coach with Claude Fable 5.2 — every · 2026-11-15