Skills
skills/inference-sh/skills/ai-music-generation/Gen Agent Trust Hub

ai-music-generation

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: All tool operations are performed through the official infsh CLI, which is the intended interface for the inference-sh platform. These commands are used legitimately for music generation tasks within the vendor's ecosystem.
  • [PROMPT_INJECTION]: The skill allows users to provide custom prompts and lyrics for audio generation. While this represents a surface for indirect prompt injection, it is considered low risk and standard for this type of AI tool. Ingestion points: user-defined JSON inputs in SKILL.md. Boundary markers: JSON object keys (prompt, lyrics). Capability inventory: execution of the infsh CLI. Sanitization: processed by the vendor's specialized audio models.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 06:55 AM