media-ytdlp

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

• Third-party content exposure detected (high risk: 0.75). At runtime, the wrapper script passes the user-supplied --url/args.url into yt-dlp (e.g., ytdlp ... -F URL / ytdlp ... URL), and yt-dlp fetches and parses page/manifest content from that URL (public web content authored by outsiders), which becomes readable text/metadata in the LLM context via the agent’s normal tool-output handling.