apex-27

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 run Membrane actions and to use "membrane request CONNECTION_ID /path/to/endpoint" (the "Proxy requests" and "Running actions" sections) which fetches data from the external Apex 27 API (third-party, potentially user-generated/untrusted content) and the agent is expected to read and act on those responses, so returned content could influence subsequent tool use.