find-skills

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a bridge to the open agent skills ecosystem, using the npx skills package manager to extend agent capabilities.
  • [COMMAND_EXECUTION]: The skill provides instructions for running command-line tools such as npx skills find, npx skills add, and npx skills update. These are standard operations for the described ecosystem and do not involve unauthorized privilege escalation.
  • [EXTERNAL_DOWNLOADS]: It facilitates the installation of skills from external sources like GitHub. The documentation specifically highlights trusted sources such as vercel-labs/agent-skills and well-known services like skills.sh.
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves and displays search results from an external database (skills.sh). While this represents a surface for indirect prompt injection from third-party skill metadata, the risk is low and inherent to the functionality of a discovery tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 06:47 PM