find-skills
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to run the 'npx skills' command for searching and installing packages. It suggests the use of the -y flag during installation, which suppresses interactive prompts and confirmation steps within the CLI tool.\n- [REMOTE_CODE_EXECUTION]: It enables the installation of external code packages from GitHub and other remote sources via the 'npx skills add' command. This is an intentional feature of the skill's discovery and extension workflow.\n- [EXTERNAL_DOWNLOADS]: The skill identifies and fetches package information from the external registry at skills.sh and third-party GitHub repositories.\n- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection via the processing of search results from the skills registry. However, it incorporates a mandatory quality verification process that requires evaluating source reputation and install metrics to ensure security.\n
- Ingestion points: External search results returned by the 'npx skills find' command (SKILL.md).\n
- Boundary markers: No explicit delimiter usage is specified for separating external data from agent instructions.\n
- Capability inventory: Execution of shell commands via npx and package installation capabilities (SKILL.md).\n
- Sanitization: Employs verification guidelines (Step 4) where the agent must check install counts, star counts, and author reputation before recommending or installing a skill.
Audit Metadata