beehiiv

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 proxy requests against the Beehiiv API (e.g., "list-posts", "get-post" and "membrane request CONNECTION_ID /path/to/endpoint"), which fetches user-generated newsletter/subscription content from a third-party service that the agent would read and could influence subsequent actions.