The Agent Skills Directory

[SAFE]: The skill consists of markdown-based workflows and a utility script for audio filtering. No malicious patterns, obfuscation, or unauthorized data access were identified during analysis.
[COMMAND_EXECUTION]: The provided scripts/voice_preset.py utility invokes the ffmpeg tool to apply audio presets. The implementation uses subprocess.run with a list of arguments and defaults to shell=False, which is a secure method for executing local binaries that prevents shell injection vulnerabilities.
[DATA_EXFILTRATION]: The documentation references the use of the ELEVENLABS_API_KEY environment variable. This aligns with secure development practices by delegating secret management to the environment rather than hardcoding credentials or accessing unauthorized locations.

ai-voice-humanizer