kvk

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

• Third-party content exposure detected (high risk: 0.70). The skill uses Membrane to proxy requests to the public KVK API (see "membrane request CONNECTION_ID /path/to/endpoint" and the KVK official docs link), meaning the agent will fetch and interpret third-party public company data as part of its workflow, which could contain untrusted/user-provided text that influences subsequent actions.