audiocraft-audio-generation
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the
audiocraftlibrary directly from Facebook Research's official GitHub repository (github.com/facebookresearch/audiocraft.git). This is a trusted source for the technology described. - [COMMAND_EXECUTION]: Includes standard command-line instructions for environment setup, such as installing dependencies via
pipand running training experiments using thedorasearch tool. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted natural language descriptions from users (e.g., in the Gradio and FastAPI examples) to generate audio content.
- Ingestion points: User-provided text strings in the
descriptionsvariable and API request bodies (SKILL.md,references/advanced-usage.md). - Boundary markers: None identified; the skill directly interpolates raw user strings into model generation calls.
- Capability inventory: Includes file system writes (saving .wav files) and network operations (FastAPI/Gradio server hosting).
- Sanitization: No validation or filtering of input strings is performed before processing.
Audit Metadata