find-skills

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is designed to fetch and install third-party code packages via the npx skills tool. It specifically references well-known technology organizations such as Vercel and Anthropic as trusted sources for these downloads.
  • [COMMAND_EXECUTION]: Uses the shell to execute package management commands, including searching (find), installing (add), and updating skills. It includes instructions to require user confirmation before performing installations.
  • [PROMPT_INJECTION]: Contains a surface for indirect prompt injection as it processes untrusted data from an external skill registry. * Ingestion points: External data from npx skills find and skills.sh search results enter the agent context. * Boundary markers: No explicit delimiters are used to wrap the external search results. * Capability inventory: The agent can execute shell commands and install software globally via the skills CLI. * Sanitization: Relies on manual AI verification of package popularity and source credibility rather than automated programmatic filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 05:56 AM