find-skills

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to download and run the skills utility. It references installation sources from well-known repositories, including Vercel Labs' official GitHub organization.
  • [COMMAND_EXECUTION]: The skill provides instructions for executing shell commands to find and add new capabilities. It suggests using the -y flag to automate installation processes.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes external data from a skill registry to influence agent actions:
  • Ingestion points: External package metadata ingested via the output of npx skills find (SKILL.md)
  • Boundary markers: No explicit markers or "ignore instructions" warnings are applied to the external search results
  • Capability inventory: The skill can install and execute code via npx skills add (SKILL.md)
  • Sanitization: Search results from the registry are presented and processed without explicit sanitization or validation of the remote metadata
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:30 PM
Security Audit — agent-trust-hub — find-skills