Skills
Navigate
skills/rafaelgrasa/skills/ai-voice-cloning/Gen Agent Trust Hub

ai-voice-cloning

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the infsh CLI to execute tasks on the inference.sh platform. This tool is explicitly listed in the allowed-tools configuration, restricting usage to defined parameters.
  • [EXTERNAL_DOWNLOADS]: References installation instructions from the inference-sh organization's public GitHub repository. This is a standard reference for setting up the required environment.
  • [DATA_EXFILTRATION]: Transmits user-provided text content to the inference.sh service via the CLI. This is the primary function of the skill and occurs within the scope of the platform's intended service.
  • [PROMPT_INJECTION]: The skill processes untrusted input data for voice generation, presenting a surface for indirect prompt injection.
  • Ingestion points: The --input JSON object in SKILL.md examples.
  • Boundary markers: Input data is wrapped in JSON keys (e.g., "prompt", "text").
  • Capability inventory: Subprocess execution of the infsh tool.
  • Sanitization: Not explicitly present in the markdown documentation, though standard for CLI input handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:26 PM