voicebox-voice-mcp-agent

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download the Voicebox source code from a GitHub repository and fetch installation scripts for the Bun and Rust development environments.\n- [REMOTE_CODE_EXECUTION]: The documentation includes standard commands to pipe remote installation scripts from well-known services (bun.sh and rustup.rs) directly to shell interpreters for environment setup.\n- [COMMAND_EXECUTION]: The skill directs the execution of shell commands for project initialization, building, and development workflows using the just task runner and the bun package manager.\n- [PRIVILEGE_ESCALATION]: The setup instructions for Linux users include standard commands for installing the FFmpeg dependency using the system package manager with sudo privileges.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and audio data for voice synthesis and cloning. 1. Ingestion points: Text strings passed to synthesis tools and audio file paths provided for cloning. 2. Boundary markers: None identified in the instructional content. 3. Capability inventory: The skill has capabilities for native file access and network requests via the Tauri platform and Bun runtime. 4. Sanitization: No specific sanitization or validation logic for external inputs is mentioned in the provided documentation.\n- [DYNAMIC_EXECUTION]: The skill uses the Tauri framework to facilitate the dynamic invocation of native Rust backend logic from its TypeScript-based frontend.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 10:44 PM
Security Audit — agent-trust-hub — voicebox-voice-mcp-agent