uservoice

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs using the Membrane proxy and actions (e.g., "membrane request CONNECTION_ID /path/to/endpoint" and action list/run) to fetch UserVoice data such as suggestions, comments and forum posts, which are user-generated, untrusted third‑party content that the agent would read and could materially influence its actions.