doc-research

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (SKILL.md steps 3-4) explicitly uses WebSearch and the web2md.ts CLI to fetch and convert public URLs (see references/web2md.md), so the agent will ingest untrusted third-party web content that could contain adversarial instructions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill calls the web2md.ts CLI which uses the Cloudflare Browser Rendering API to fetch arbitrary external pages at runtime (e.g., https://developers.cloudflare.com/) and converts that fetched content to Markdown which is then injected into the agent's context and can directly control prompts/instructions.