find-skills
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill is designed to download and install external code via 'npx skills add'. It specifically references content from well-known and trusted sources like Vercel Labs.- [COMMAND_EXECUTION]: Orchestrates the execution of shell commands ('npx skills') to search for, install, and update modular agent packages.- [DATA_EXPOSURE_AND_EXFILTRATION]: Performs network operations to interact with the 'skills.sh' registry and GitHub for skill discovery and retrieval.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the 'npx skills find' search results (skill descriptions/names). While this presents a surface for indirect prompt injection if a malicious skill has a deceptive description, the impact is limited to recommending the installation of that skill.
Audit Metadata