ai-sdk

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs runtime fetching of public web content (ai-sdk.dev docs via https://ai-sdk.dev/api/search-docs and subsequent .md URLs) which would be ingested as readable text for the agent to use.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs runtime fetching of external docs and model lists (e.g., https://ai-gateway.vercel.sh/v1/models and https://ai-sdk.dev/api/search-docs?q=your_query / https://ai-sdk.dev/docs/agents/building-agents.md), which are fetched at runtime and directly influence model selection and produced code/prompt decisions.