insightoai

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md shows the agent uses Membrane to query Insighto.ai (e.g., "Get Conversation Transcript", "List Conversations", and proxy requests) which will fetch user-generated/untrusted content (conversation transcripts, contacts, feedback) from the third-party Insighto.ai service that the agent is expected to read and could materially influence subsequent actions.