Skills
skills/inference-sh-8/skills/ai-music-generation/Gen Agent Trust Hub

ai-music-generation

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to run infsh CLI commands. Access is restricted via the allowed-tools frontmatter to only permit commands beginning with infsh *, which limits the risk of arbitrary command execution.
  • [EXTERNAL_DOWNLOADS]: The skill references installation scripts and documentation hosted on the vendor's official GitHub repository (inference-sh/skills) and website (inference.sh). These are identified as legitimate vendor-owned resources.
  • [PROMPT_INJECTION]: The skill accepts user-provided strings for prompts and lyrics which are interpolated into CLI command arguments, representing an indirect injection surface.
  • Ingestion points: User-defined prompts and lyrics in infsh app run payloads within SKILL.md examples.
  • Boundary markers: Inputs are passed within structured JSON payloads, which provides structural delimitation.
  • Capability inventory: The agent can only execute infsh commands and lacks general file-write or unrestricted network access tools.
  • Sanitization: Input sanitization is deferred to the infsh CLI utility.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 01:59 PM
Security Audit — agent-trust-hub — ai-music-generation