SkillsMCP servers

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.

Stars
27,188
+7 days
—
Forks
2,419
Language
Python
License
Apache-2.0
Last push
20 Nov 2026

Summary

Summary written by AI from the README; it may be imprecise.

This security scanner audits AI agent skills and MCP servers to detect vulnerabilities, malicious patterns, and prompt injection before installation. It combines static analysis using AST rules and YARA signatures with optional LLM semantic evaluation and live CVE checks against OSV.dev. Developers can run it as a CLI tool, Docker container, or MCP server to gate skill installations directly inside Claude Code.

Use it when

  • Vet third-party AI agent skills or MCP servers before installing them into Claude Code
  • Gate skill installations automatically inside agent workflows using the MCP server interface
  • Scan local skill directories or Git repositories for prompt injection and supply-chain risks

Install

mcp
claude mcp add skillspector -- skillspector mcp
pip
uv tool install --python 3.13 git+https://github.com/NVIDIA/skillspector.git
pip
uv tool install --python 3.13 'skillspector[mcp] @ git+https://github.com/NVIDIA/skillspector.git'

Commands quoted from the repo's README. Read the README before running them.

Needs a token or API key.

Stars over 90 days

Collecting: 2 of 90 days so far. The chart appears after 7 days.

Type to search the reference, templates and lessons.