Skill Vetter
Skill Vetter analyzes your AI agent’s outputs to identify potential issues like hallucinations or biases, ensuring more reliable and trustworthy responses.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add skill-vetter npx -- -y @trustedskills/skill-vetter
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"skill-vetter": {
"command": "npx",
"args": [
"-y",
"@trustedskills/skill-vetter"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The skill-vetter tool evaluates and verifies the quality, safety, and reliability of AI agent skills before they are deployed or used. It checks for potential vulnerabilities, ensures compliance with best practices, and confirms that the skill functions as intended.
When to use it
- Before deploying a new AI agent skill in production to ensure it meets security standards
- When reviewing third-party skills from external repositories or marketplaces
- During development to catch issues early in the testing phase
Key capabilities
- Security vulnerability scanning
- Compliance verification with industry standards
- Functional correctness checks
Example prompts
- "Analyze this AI agent skill for potential security risks."
- "Verify if this skill complies with data privacy regulations."
- "Check whether this skill performs its intended function correctly."
Tips & gotchas
- Ensure the tool has access to up-to-date vulnerability databases for accurate results.
- Some advanced checks may require additional configuration or integration with external systems.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.