ffmpeg-docs

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

• Third-party content exposure detected (high risk: 0.85). The skill fetches and text-extracts outsider-authored free text from ffmpeg.org at runtime (public web pages under PAGES via urllib.request.urlopen in load_or_fetch() → TextExtractor.feed()), which is then printed/returned and can be included in the agent’s LLM context.