{"name":"SkillSpector: NVIDIA's Security Scanner for AI Agent Skills","description":"SkillSpector is a critical security scanner developed by NVIDIA for AI agent skills. It identifies vulnerabilities, malicious patterns, and various security risks, including prompt injection and data exfiltration, in skills for platforms like Claude Code, Codex, and MCP. This tool empowers developers and users to ensure the safety and integrity of AI agent environments before skill installation.","github":"https://github.com/NVIDIA/SkillSpector","url":"https://osrepos.com/repo/nvidia-skillspector","source":"osrepos.com","sourceDescription":"This repository profile is provided by osrepos.com, an open source repository discovery platform.","repositoryProfile":"https://osrepos.com/repo/nvidia-skillspector","generatedFor":"open source discovery and AI-assisted research","markdown":"https://osrepos.com/repo/nvidia-skillspector.md","json":"https://osrepos.com/repo/nvidia-skillspector.json","topics":["agent-security","ai-security","prompt-injection","supply-chain-security","security-scanner","python","ai-agents","vulnerability-detection"],"keywords":["agent-security","ai-security","prompt-injection","supply-chain-security","security-scanner","python","ai-agents","vulnerability-detection"],"stars":null,"summary":"SkillSpector is a critical security scanner developed by NVIDIA for AI agent skills. It identifies vulnerabilities, malicious patterns, and various security risks, including prompt injection and data exfiltration, in skills for platforms like Claude Code, Codex, and MCP. This tool empowers developers and users to ensure the safety and integrity of AI agent environments before skill installation.","content":"## Introduction\n\nAI agent skills, utilized by platforms such as Claude Code, Codex CLI, and Gemini CLI, often execute with implicit trust and minimal vetting. Research indicates a significant number of these skills contain vulnerabilities, with a notable percentage showing likely malicious intent. SkillSpector addresses this critical concern by providing a robust security scanner that helps answer the fundamental question: \"Is this skill safe to install?\"\n\nDeveloped by NVIDIA, SkillSpector is an integral part of the NVIDIA Verified Skills pipeline, which meticulously scans, evaluates, and signs agent skills prior to their publication. Skills that successfully pass these rigorous checks are then made available in the NVIDIA skills catalog.\n\n## Why Use SkillSpector & Key Features\n\nThe proliferation of AI agent skills necessitates a strong security posture. SkillSpector is designed to proactively detect a wide array of threats, safeguarding your AI agents from compromise. Here are its key features:\n\n*   **Multi-format Input**: Scan various sources including Git repositories, URLs, zip files, local directories, or single files.\n*   **Extensive Vulnerability Patterns**: Detects 71 vulnerability patterns across 17 categories, covering prompt injection, data exfiltration, privilege escalation, supply chain risks, excessive agency, output handling, system prompt leakage, memory poisoning, tool misuse, rogue agent, anti-refusal, trigger abuse, dangerous code (AST), taint tracking, YARA signatures, MCP least privilege, and MCP tool poisoning.\n*   **Two-Stage Analysis**: Employs a fast static analysis for broad detection, complemented by an optional LLM semantic evaluation to refine findings and reduce false positives.\n*   **Live Vulnerability Lookups**: Integrates with OSV.dev to provide real-time CVE data for dependencies, with an automatic offline fallback.\n*   **Multiple Output Formats**: Generates reports in various formats, including Terminal, JSON, Markdown, and SARIF, for diverse integration needs.\n*   **Risk Scoring**: Provides a clear 0-100 risk score with severity labels and actionable recommendations, helping you quickly assess the threat level.\n*   **Baseline / False-Positive Suppression**: Allows you to accept known findings via glob-rules or fingerprint baselines, ensuring that subsequent scans highlight only new issues.\n\n## Installation\n\nSkillSpector offers flexible installation options to suit your workflow.\n\n### Quick Install with `uv` (CLI-only)\n\nbash\nuv tool install git+https://github.com/NVIDIA/skillspector.git\n# Update later: uv tool update skillspector\n\n# If you plan to run 'skillspector mcp', install the MCP extra:\nuv tool install 'skillspector[mcp] @ git+https://github.com/NVIDIA/skillspector.git'\n\n\n### From Source\n\nbash\n# Clone the repository\ngit clone https://github.com/NVIDIA/skillspector.git\ncd skillspector\n\n# Create and activate virtual environment\nuv venv .venv && source .venv/bin/activate\n# or: python3 -m venv .venv && source .venv/bin/activate\n\n# Install for production use\nmake install\n\n# Or install with development dependencies\nmake install-dev\n\n\n### Docker (no Python required)\n\nbash\n# Build the image\nmake docker-build\n# or: docker build -t skillspector .\n\n# Scan a local directory\ndocker run --rm -v \"$PWD:/scan\" skillspector scan ./my-skill/ --no-llm\n\n# Scan with LLM analysis (using a .env file for credentials)\ncat > .env <<'EOF'\nSKILLSPECTOR_PROVIDER=anthropic\nANTHROPIC_API_KEY=sk-ant-...\nEOF\ndocker run --rm \\\n  -v \"$PWD:/scan\" \\\n  --env-file .env \\\n  skillspector scan ./my-skill/\n\n\n## Examples\n\n### Basic Usage\n\nbash\n# Scan a local skill directory\nskillspector scan ./my-skill/\n\n# Scan a single SKILL.md file\nskillspector scan ./SKILL.md\n\n# Scan a Git repository\nskillspector scan https://github.com/user/my-skill\n\n# Scan a zip file\nskillspector scan ./my-skill.zip\n\n\n### Output Formats\n\nbash\n# Terminal output (default)\nskillspector scan ./my-skill/\n\n# JSON output for machine readability\nskillspector scan ./my-skill/ --format json --output report.json\n\n# Markdown output for documentation\nskillspector scan ./my-skill/ --format markdown --output report.md\n\n# SARIF output for CI/CD integration\nskillspector scan ./my-skill/ --format sarif --output report.sarif\n\n\n### LLM Analysis Configuration\n\nFor enhanced semantic analysis, SkillSpector can integrate with various LLM providers. For example, using OpenAI:\n\nbash\nexport SKILLSPECTOR_PROVIDER=openai\nexport OPENAI_API_KEY=sk-...\nskillspector scan ./my-skill/\n\n\nOr to skip LLM analysis for faster, static-only scans:\n\nbash\nskillspector scan ./my-skill/ --no-llm\n\n\n## Links\n\n*   **GitHub Repository**: [NVIDIA/SkillSpector](https://github.com/NVIDIA/SkillSpector){:target=\"_blank\"}\n*   **Official Documentation**: [Scan agent skills before installation](https://docs.nvidia.com/skills/scanning-agent-skills){:target=\"_blank\"}\n*   **License**: [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0){:target=\"_blank\"}","metrics":{"detailViews":0,"githubClicks":0},"dates":{"published":null,"modified":"2026-09-11T20:33:50.000Z"}}