voice-clone-narration
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for local-first operation, ensuring that audio samples and generated narrations remain on the user's filesystem at a dedicated home directory (~/.voice-clone-narration/).
- [COMMAND_EXECUTION]: The skill uses
ffmpegfor audio processing viasubprocess.runand shell scripts. The Python implementation correctly uses a list of arguments rather than a shell string, effectively mitigating command injection risks. - [EXTERNAL_DOWNLOADS]: The skill downloads model weights from Hugging Face repositories (e.g., mlx-community/chatterbox-fp16) during the first run. These downloads target a well-known, trusted platform and are necessary for the skill's stated purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text for speech synthesis via
narrate.py. While this presents a potential injection surface, the risk is limited to the text-to-speech output, and the skill includes documentation on safety and consent for biometric data usage.
Audit Metadata