skill-security-auditor - Audit AI agent skills for pre-installation security risks
Scans AI agent skills for prompt injection, dangerous code, data exfiltration, dependency risks, boundary violations, and obfuscation.
Tags
Updated: 2026-10-07Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Detect prompt injection
- Scan dangerous code patterns
- Analyze dependency supply chains
- Check filesystem boundaries
- Detect obfuscation
- Produce audit verdicts
- Provide remediation guidance
Inputs
- Skill markdown files
- Skill code files
- Dependency manifests
- Skill directories
- Git repositories
Outputs
- Structured audit report
- PASS/WARN/FAIL verdict
- Severity-categorized findings
- Remediation guidance