ralph

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and use external documentation via ctx7 (SKILL.md "Verify external docs" step) and the PRD/issue text requires downloading goldens from the Confident AI dataset (EvaluationDataset.pull alias="chat-gold"), so the agent will ingest untrusted third‑party content that can materially influence API calls, tests, and subsequent actions.