action-network

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 live, user-generated Action Network content (e.g., via membrane action run for events/petitions/signatures and membrane request CONNECTION_ID /path/to/endpoint proxying to the Action Network API), so untrusted third‑party text could be ingested and influence subsequent actions.