codereadr

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly uses Membrane to call the external CodeREADr API (e.g., "membrane action run" and "membrane request CONNECTION_ID /path/to/endpoint" and actions like "list-scans" / "list-database-values"), which ingests user-generated/untrusted scan and database content from a third-party service that the agent is expected to read and can influence subsequent actions (create/edit/delete), creating a clear avenue for indirect prompt injection.