swell

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md instructs using Membrane to proxy requests to the Swell API (e.g., "membrane request CONNECTION_ID /path/to/endpoint") and to run actions that read Swell data (customer reviews, surveys, web chat), which are untrusted/user-generated third‑party contents that the agent is expected to read and could materially influence subsequent actions.