skill-auditor

Scan agent skill directories for security risks and generate SKILL_AUDIT.md verdicts.

9|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/Montimage/skills --skill skill-auditor-montimage
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-auditor
Source: https://github.com/Montimage/skills/tree/main/skills/skill-auditor
Command: npx skills add https://github.com/Montimage/skills --skill skill-auditor-montimage

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Analyze agent skill directories for security risks and safe installation.

Core Features & Use Cases

  • Automated security scanning of skill directories for dangerous patterns, prompt injection risks, and data exfiltration indicators.
  • Generates a comprehensive SKILL_AUDIT.md report with an install/reject verdict.
  • Intercepts installation decisions to ensure safe onboarding of third-party skills.

Quick Start

Audit a target skill directory to obtain a safety verdict and actionable findings.

Frequently Asked Questions about skill-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit agent skills for security risks before installation?

To audit agent skills for security risks, run the built-in script to scan target directories, parse SKILL.md frontmatter, and generate a structured SKILL_AUDIT.md report with a clear install or reject verdict.

What security risks are detected during an agent skill audit?

An agent skill audit detects dangerous patterns, prompt injection risks, and data exfiltration indicators by scanning skill directories and parsing SKILL.md frontmatter to produce a structured safety report.

Can I audit GitHub-hosted agent skill packages for safe installation?

Yes, you can audit GitHub-hosted agent skill packages. The security scanning applies to both local skill directories and GitHub-hosted packages, covering audit scenarios from single-skill validation to bulk reviews.

How do I generate a security compliance report for a local skill directory?

Generate a security compliance report by running the built-in scanning script against a local skill directory. It parses frontmatter and outputs a structured SKILL_AUDIT.md file containing actionable findings and an install or reject verdict.

Does the skill audit support bulk reviews of multiple agent skills?

Yes, the skill audit supports bulk reviews of multiple agent skills. It covers audit scenarios from single-skill validation to bulk reviews of GitHub-hosted packages and local directories.

What limitations exist when scanning agent skills for prompt injection risks?

Scanning limitations depend on parsing SKILL.md frontmatter and built-in script patterns. Auditing targets local directories and GitHub-hosted packages, so skills lacking standard SKILL.md files may not produce a complete SKILL_AUDIT.md verdict.