audiocraft-audio-generation
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions include downloading the AudioCraft library and pre-trained models from official repositories on GitHub and HuggingFace. These references target established and well-known technology organizations.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides code to generate audio from text descriptions, creating an ingestion surface for untrusted data.\n
- Ingestion points: Input descriptions used in
SKILL.mdand the prompt variables in the FastAPI and Gradio examples inreferences/advanced-usage.md.\n - Boundary markers: No specific boundary markers or instructions to ignore embedded commands are included in the prompt interpolation.\n
- Capability inventory: The skill facilitates audio generation using pre-trained models and file-writing operations via
torchaudioandscipy.\n - Sanitization: Input prompts are processed without explicit sanitization or filtering logic.\n- [COMMAND_EXECUTION]: The documentation includes standard shell commands for system dependency installation (e.g.,
ffmpeg) and development environment configuration (e.g., creating YAML files).
Audit Metadata