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 guidelines and related skill definitions from the inference-sh GitHub repository and the official inference.sh domain. These are standard documentation and platform references.
  • [COMMAND_EXECUTION]: Instructs the agent to use the belt CLI tool for authenticating and running remote voice transformation applications. The allowed-tools configuration correctly scopes bash usage to the belt command prefix.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external audio URLs. 1. Ingestion points: The audio field in the JSON input for the belt command in SKILL.md. 2. Boundary markers: The input is structured within a JSON object. 3. Capability inventory: The skill uses the Bash tool to execute belt. 4. Sanitization: No explicit sanitization of the URL is mentioned, but the content is processed by an audio API for signal transformation, not by the LLM as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:19 PM
Security Audit — agent-trust-hub — elevenlabs-voice-changer