Show HN: Audionaut – Open-Source Multitrack Audio Editor with MCP Support for AI Agents

AI Agent Integration via MCP

Audionaut is a free, open-source desktop application for audio editing that integrates with AI agents through the Model Context Protocol (MCP). This allows agents like Claude to directly manipulate audio sessions, performing tasks such as cutting songs, splitting clips, and setting crossfades. Each edit performed by an agent is recorded as a single undo step within the application. Users can connect Claude to Audionaut using the command claude mcp add audionaut -- npx -y audionaut-mcp.

Core Features and Architecture

Written in C++ on the JUCE framework, Audionaut runs natively on Windows, macOS, and Linux. It is designed as a lightweight alternative to full Digital Audio Workstations (DAWs), offering precise cutting, per-track playlists, and multi-channel support. Key technical capabilities include:

• Stem Separation: Utilizes demucs.cpp (a C++ port of Meta's Demucs) to separate audio into Drums, Bass, Other, and Vocals tracks.
• Audio Analysis: Links against Essentia for features like BIC segmentation, onset detection, and beat tracking.
• Headless CLI: Includes audionaut-cli for scripts and CI, supporting commands for creating projects, importing audio, analyzing, auto-editing, and exporting with machine-readable JSON output.

Licensing and Availability

The project is dual-licensed under GPL3 (or later) and a commercial license. It is available on GitHub, with documentation covering installation, building from source, and detailed CLI usage for agent workflows.