seeking-skills

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill implements functionality to search for and download content from external GitHub repositories using the /skill-seeker:seek and /skill-seeker:install commands.
  • [REMOTE_CODE_EXECUTION]: The /skill-seeker:install command facilitates the installation of external logic from community repositories, allowing for the introduction and execution of unverified third-party code within the agent's environment.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of unverified third-party skill content.
  • Ingestion points: External GitHub repositories accessed via search and install commands (SKILL.md).
  • Boundary markers: None identified in the provided documentation to distinguish between core instructions and fetched skill data.
  • Capability inventory: The skill allows for the management, loading, and execution of functional extensions (skills) sourced from external environments.
  • Sanitization: No explicit sanitization or verification of the fetched skill content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 06:41 PM
Security Audit — agent-trust-hub — seeking-skills