partnerize

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 using membrane to proxy arbitrary Partnerize API endpoints (e.g., "membrane request CONNECTION_ID /path/to/endpoint") and to run actions (membrane action run), so the agent will fetch and read third‑party Partnerize responses that can directly influence subsequent actions and decisions.