ai-sdk-development

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Provider Tools" and "Tools" sections explicitly show adding WebSearch and WebFetch provider tools (e.g., "new WebFetch", "(new WebSearch)->max(5)->allow(['laravel.com'])"), which enables the agent to fetch and read open/public web pages or arbitrary URLs and thus to consume untrusted, third‑party content that can influence its actions.