find-skills

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the npx skills CLI tool to perform various operations, including searching for skills (find), checking for updates (check), and initializing new skills (init).
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of external code packages from GitHub and other remote sources using the npx skills add command. These operations target established ecosystems and include references to trusted organizations.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The agent processes and presents output from the npx skills find command, which contains descriptions and metadata from external skill packages.
  • Boundary markers: There are no explicit delimiters or warnings to ignore instructions that might be embedded in the search results.
  • Capability inventory: The skill possesses the ability to execute shell commands and install software packages.
  • Sanitization: The skill does not describe specific methods for sanitizing or validating the metadata retrieved from the skill registry before it is presented to the user or agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 02:58 PM
Security Audit — agent-trust-hub — find-skills