inworld-ai

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

• Third-party content exposure detected (high risk: 0.90). The skill instructs the agent to fetch and interact with Inworld AI data via Membrane (e.g., using "membrane action run" and the "membrane request CONNECTION_ID /path/to/endpoint" proxy), which pulls potentially user-generated/untrusted third-party content that the agent is expected to read and act on as part of its workflow.