elevenlabs-voice-changer

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions and related tools hosted on the official inference-sh GitHub repository. It also suggests installing a CLI component using npx. These are functional dependencies required for the skill's operation.
  • [COMMAND_EXECUTION]: The skill uses the belt CLI to run voice transformation tasks. The allowed-tools configuration in the frontmatter restricts shell access exclusively to the belt command, which follows the principle of least privilege.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external audio files via URLs, which represents a potential attack surface for indirect injection if the processing engine were vulnerable to audio-embedded instructions.
  • Ingestion points: The audio input parameter in the belt app run commands (SKILL.md).
  • Boundary markers: None present.
  • Capability inventory: Restricted shell execution via Bash(belt *) (SKILL.md).
  • Sanitization: None identified; the skill relies on the remote platform (inference.sh) for audio processing and sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — elevenlabs-voice-changer