ai-voice-cloning

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill takes text inputs such as 'prompt' and 'text' from users to generate speech. This represents a potential surface for indirect prompt injection if the output audio or the generated text is later processed by an AI agent that might interpret embedded instructions.
  • Ingestion points: 'prompt' and 'text' fields in the JSON input for the belt CLI (found in SKILL.md examples).
  • Boundary markers: None present to delimit user input from instructions.
  • Capability inventory: The skill utilizes the 'Bash(belt *)' tool to execute inference commands.
  • Sanitization: No explicit sanitization or validation of the input text is described within the skill instructions.
  • [EXTERNAL_DOWNLOADS]: The skill references and encourages the installation of the 'belt-sh/cli' skill and points to an external GitHub repository (inference-sh/skills) for installation instructions. These external references are standard for integrating with the required inference platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 05:09 PM
Security Audit — agent-trust-hub — ai-voice-cloning