elevenlabs-dubbing

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text via the user-supplied --input JSON fields like audio (a URL) and captions/transcript (free-form text), which the LLM/CLI then uses after fetching/processing external content (e.g., public web video/audio and its extracted text).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes the inference.sh platform via the belt CLI (e.g., belt app run elevenlabs/dubbing), which executes remote app code on https://inference.sh at runtime and thus relies on external code controlling the agent's behavior.