graphy

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to fetch and proxy content from the Graphy API via Membrane (see "Proxy requests" and the "membrane action run" / "membrane request CONNECTION_ID /path/to/endpoint" sections), which can return untrusted, user-generated social media content that the agent is expected to read and could alter subsequent actions.