ydc-ai-sdk-integration

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly invokes You.com web tools (youSearch, youResearch, youContents) to fetch raw content from arbitrary public websites—see the SKILL.md "Available Tools" and "Security" sections and examples (assets/path-a-generate.ts, assets/path-b-stream.ts)—and those tool results are ingested into the model's context, creating a clear indirect prompt-injection surface.