ai-voice-cloning

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install external dependencies such as the belt CLI via npx skills add belt-sh/cli and refers to an installation instruction script hosted at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.
  • [COMMAND_EXECUTION]: The skill uses the belt CLI tool to run various AI voice models through shell commands (e.g., belt app run inworld/text-to-speech-2).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text to generate audio, which presents a surface for indirect prompt injection.
  • Ingestion points: User input provided via the prompt and text fields in JSON payloads for the belt command.
  • Boundary markers: Commands use structured JSON to encapsulate user text, which provides basic delimitation.
  • Capability inventory: The skill uses Bash(belt *) to execute remote applications for audio synthesis and media merging.
  • Sanitization: No specific sanitization or filtering of the input text is mentioned in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 01:13 PM
Security Audit — agent-trust-hub — ai-voice-cloning