typeflowai

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 fetch and run actions against the TypeflowAI service (e.g., "membrane action run ..." and "membrane request CONNECTION_ID /path/to/endpoint"), which retrieves third‑party TypeflowAI data (potentially user-generated) that the agent is expected to read and act on, enabling indirect prompt injection.