html-skills-listen

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text can enter the LLM context via the runtime path where server.js forwards arbitrary HTTP POST bodies (content: body) to stdout as JSON-RPC notifications, and html-skills-listen arms Monitor to tail that stdout and surface submissions to the agent.