jotform

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests user-generated Jotform data via Membrane actions (e.g., "list-form-submissions", "get-form-questions") and via the "membrane request" proxy to the Jotform API, so untrusted third-party form/submission content would be read by the agent and could influence subsequent actions (create/update/delete) and decisions.