anycap-cli

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly fetches and ingests arbitrary public web and user-provided URLs (e.g., "anycap crawl " in references/crawl.md, "anycap search --query/--prompt" in references/search.md, and actions that accept --url like "anycap actions image-read --url https://..."/video-read/audio-read in references/actions.md) and the agent is expected to read and synthesize that content (grounding search, crawl -> markdown, actions -> content) which can materially influence subsequent decisions or tool use.