acestep
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content such as music prompts and lyrics that are passed as arguments to a local python tool.
- Ingestion points: The skill accepts arbitrary text strings for the --prompt and --lyrics parameters, and loads external configuration or data via the --brand, --reference, and --input parameters in SKILL.md.
- Boundary markers: Examples demonstrate basic shell quoting for arguments but lack explicit instructions to sanitize embedded commands or instructions within the user-provided text.
- Capability inventory: The skill utilizes subprocess calls to tools/music_gen.py across multiple functional areas in SKILL.md.
- Sanitization: No input validation, escaping mechanisms, or schema enforcement are described for the user-provided text segments or referenced files.
Audit Metadata