chat-with-anyone

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

• Third-party content exposure detected (low risk: 0.10). The workflow downloads YouTube/Bilibili videos and processes their subtitle files (SRT) into the LLM context or execution pipeline, allowing outsider-authored text from video captions/subtitles to influence the agent.

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

• Potentially malicious external URL detected (high risk: 0.70). An operational dependency on noiz.ai is present via DEFAULT_BASE_URL and API calls during runtime, but noiz.ai lacks established first-party provenance or verification, making it a weak runtime dependency.