ydc-langchain-integration

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

• Third-party content exposure detected (high risk: 1.00). The skill integrates the youSearch and youContents tools (SKILL.md Integration Example and Security sections, plus assets/reference.ts) which fetch raw content from arbitrary public websites and pass those results into createAgent so the agent reads and acts on that untrusted web content, creating an indirect prompt-injection surface.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill uses the youContents tool to fetch arbitrary external webpages at runtime (e.g., the example URL https://example.com) whose raw HTML/markdown is injected into the model context as tool results, creating a high-confidence indirect prompt-injection surface.