acestep
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface as it ingests and processes untrusted data from multiple sources to generate audio content.
- Ingestion points: Text prompts provided via the
--promptargument, song lyrics provided via the--lyricsargument, and reference audio files provided via the--referenceand--inputarguments inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions to the model to ignore potential commands embedded within the user-provided lyrics or prompts.
- Capability inventory: The skill executes a local script
tools/music_gen.py, which is capable of writing audio files to the disk (--output) and communicating with RunPod serverless endpoints. - Sanitization: There is no evidence of input sanitization or filtering for the text-based prompts or lyrics before they are processed by the generative tool.
Audit Metadata