hol-guard
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local CLI commands such as hol-guard and plugin-scanner to manage AI harness security and perform artifact analysis.
- [EXTERNAL_DOWNLOADS]: The skill recommends installing security management and scanning tools via pipx (e.g., pipx install hol-guard, pipx install plugin-scanner).
- [INDIRECT_PROMPT_INJECTION]: The skill functions as a security scanner for external plugins, skills, and marketplace packages, which involves processing potentially untrusted data.
- Ingestion points: File paths provided to the plugin-scanner verify command for external repositories and packages.
- Boundary markers: The instructions do not explicitly define prompt boundary markers for the scanner output.
- Capability inventory: The skill utilizes file system read capabilities and execution capabilities for authorized security tools.
- Sanitization: The skill relies on the internal logic of the plugin-scanner tool for content validation rather than defining explicit sanitization in the prompt instructions.
Audit Metadata