skill-finder

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly searches GitHub for SKILL.md files and uses read_url_content/curl to fetch raw SKILL.md content from arbitrary public repositories, then reads/opens those files locally, which exposes the agent to untrusted, user-generated third-party content from GitHub.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches raw SKILL.md files at runtime (e.g., raw.githubusercontent.com/.../SKILL.md) and writes/loads them as agent skills, meaning remote content directly controls agent instructions and is a required runtime dependency.