acestep
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill functions by executing a local script
tools/music_gen.py. This script handles communication with various backends (acemusic.ai, Modal, RunPod) for music generation tasks such as text-to-music, stem extraction, and audio repainting. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data via the
--promptand--lyricsarguments. This data is passed to an AI model, creating a potential surface for indirect prompt injection if the source of the input is untrusted (e.g., content extracted from a website or untrusted document). - Ingestion points: Arguments passed to
tools/music_gen.py(e.g.,--prompt,--lyrics). - Boundary markers: No specific delimiters or "ignore previous instruction" warnings are visible in the markdown instructions for the tool's usage.
- Capability inventory: Subprocess execution of a Python script which likely performs network requests to external APIs.
- Sanitization: The instructions do not describe any sanitization or validation performed on the user-provided text before it is sent to the model.
Audit Metadata