goodbits

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to run Membrane actions (e.g., "membrane action run" and "membrane request CONNECTION_ID /path/to/endpoint") to fetch Goodbits data such as emails, newsletters, and subscriber content—external, user-generated third-party content that the agent will read and act on—so it can be used to inject indirect instructions.