talkjs

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md shows the agent using Membrane to run actions and proxy requests against TalkJS (e.g., "membrane action list/run" and "membrane request CONNECTION_ID /path/to/endpoint" in the Running actions and Proxy requests sections), which will fetch user-generated chat/conversation data from a third-party service that the agent is expected to read and could materially influence subsequent actions.