acestep

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill interfaces with a local script tools/music_gen.py to perform various audio tasks including generation, stem extraction, and style transfer. It facilitates this through standard shell commands with user-defined parameters.
  • [SAFE]: Credential management is handled securely by requiring users to define RUNPOD_API_KEY and RUNPOD_ACESTEP_ENDPOINT_ID in a .env file, adhering to secret management best practices.
  • [SAFE]: The skill uses localized temporary file storage (/tmp/lyrics.txt) for processing multi-line user input, which is a standard and safe method for handling transient data.
  • [PROMPT_INJECTION]: The skill processes user-provided prompts and lyrics as command arguments, creating a potential surface for indirect prompt injection. 1. Ingestion points: --prompt and --lyrics in SKILL.md. 2. Boundary markers: Absent in command-line examples. 3. Capability inventory: Subprocess execution of music_gen.py. 4. Sanitization: Not explicitly documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 12:53 PM