read-github

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

• Third-party content exposure detected (medium risk: 0.30). The skill reads GitHub repository documentation and code via gitmcp.io, which exposes the agent to outsider-authored repository contents and code.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill relies on the third-party gitmcp.io MCP service for code and documentation retrieval without established official provenance, representing a weak/unknown operational dependency.