hook-manager
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses standard shell commands to check for the presence of dependencies (jq, agent-browser, python3) and to locate the latest version of the plugin manifest file in the user's home directory.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading manifest.json from the local plugin cache to present hook options to the user.\n
- Ingestion points: manifest.json in the plugin cache directory (read in SKILL.md).\n
- Boundary markers: None explicitly defined for the manifest content display.\n
- Capability inventory: Uses the Read tool to access configs and the Write tool to update hooks-config.json.\n
- Sanitization: The skill relies on structured JSON parsing by the agent to categorize hooks.\n- [EXTERNAL_DOWNLOADS]: The README.md file references an installation command using bunx for the b-open-io/bopen-tools package, which is a vendor-owned resource.
Audit Metadata