AudioMind

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: Extensive review of the skill's source code, including the proxy implementation and CLI tools, revealed no malicious patterns, obfuscation, or safety bypass attempts. The skill operates as described, providing a unified interface for audio generation services.\n- [COMMAND_EXECUTION]: The skill performs expected local operations to facilitate audio generation and storage.\n
  • Evidence: The script tools/start_server.sh is used to launch the elevenlabs-mcp server. Additionally, cli.js writes generated audio payloads to the local file system at ~/.openclaw/workspace/tmp/audiomind for delivery to the user.\n- [EXTERNAL_DOWNLOADS]: The skill interacts with external service endpoints to process and retrieve audio data.\n
  • Evidence: Requests are made to the well-known ElevenLabs API (api.elevenlabs.io) and fal.ai. By default, the CLI communicates with a hosted proxy at audiomind-proxy.vercel.app, which is managed by the skill author, though users have the documented option to self-host this component for enhanced privacy.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 12:15 PM