slicktext

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs the agent to fetch and run actions against the external SlickText API via Membrane (e.g., "membrane request CONNECTION_ID /path/to/endpoint" and "membrane action run"), which will return untrusted/user-generated data (messages, contacts, etc.) that the agent is expected to read and could materially influence subsequent actions.