appbaseio

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md) explicitly instructs the agent to call Membrane against external Appbase.io endpoints (e.g., "membrane request CONNECTION_ID /path/to/endpoint" and "membrane action list") and to consume action results and connection metadata such as clientAction.agentInstructions, meaning the agent will fetch and interpret potentially untrusted, third‑party/user content and follow instructions returned by those external sources.