research-pipeline

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

• Third-party content exposure detected (high risk: 0.90). The skill's Phase 2 (GATHER) and scope.md explicitly require agents to use "web" source types (e.g., "Source types: web + docs" and "Save your findings to: research/{topic}/raw-{angle}.md"), causing the system to fetch and ingest public web content into raw-{angle}.md which is then read in Phase 3 and used to drive synthesis/validation and next actions, exposing the agent to untrusted third‑party content.