seeking-skills

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). This SKILL.md explicitly instructs the agent to search, browse, fetch, and install community skills from public GitHub repositories via /skill-seeker:seek, /skill-seeker:browse, and /skill-seeker:install, which ingests untrusted, user-generated third-party content that can materially change agent behavior and enable indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill exposes runtime commands that search and install code from GitHub (e.g., owner/repo, https://github.com// or git@github.com:org/repo.git) — fetched repository content would be loaded/installed at runtime and could directly control agent prompts or execute remote code.