recipe-front-adjust

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

• Third-party content exposure detected (high risk: 0.75). The workflow explicitly performs “External Resource Hearing” via external-resource-context and reads docs/project-context/external-resources.md to resolve “relevant UI sources through declared access methods,” which at runtime can fetch and ingest public/outsider web or other third-party free-text content into the LLM’s context.