⬢github Python · 22.1K ★ +445 since we first saw it · pushed 3 d ago · Apache-2.0
NVIDIA/SkillSpector
Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills before you install them.
SkillSpector is an open-source security scanner from NVIDIA for AI agent skills (Claude Code, Codex, MCP, etc.). Before you install a skill, it scans repos, URLs, zips, or files for 71 vulnerability patterns—prompt injection, data exfiltration, supply-chain risks, tool poisoning—using static analysis plus optional LLM review, and produces a 0-100 risk score with SARIF/JSON/Markdown reports.
Why now: It's newly released by NVIDIA with striking research stats (over a quarter of analyzed skills contain vulnerabilities), and it's part of NVIDIA's Verified Skills pipeline, driving interest in agent-skill supply-chain security.
Who it is for: Developers and security teams who install or publish AI agent skills and want to vet them for malicious or risky behavior.
agent-securityagent-skillsagentic-aiai-securityclaude-codemcpprompt-injectionsecurity-scannersecurity-toolssecurity-workflow
Stars over our 48 snapshots: 21.7K to 22.1K, since 11 h ago.
Where people talked about it
- ⬢github NVIDIA/SkillSpector 11 min ago
API: https://socialmediatrends-api.osmike.com/v1/repos/NVIDIA/SkillSpector