smartling

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly lets the agent fetch and proxy content from the Smartling service (see "Proxy requests" / membrane request CONNECTION_ID /path/to/endpoint and action runs), which pulls third‑party/user-provided translation strings/files that the agent is expected to read and could influence subsequent actions.