reaper

This MCP server provides tools to analyze and inspect your Reaper projects from an MCP client (like Claude Desktop). It exposes: find_reaper_projects to locate Reaper project files in a configured directory; parse_reaper_project to extract detailed project data such as tempo, tracks, FX chains, and audio items; list_installed_fx to enumerate installed plugins (filterable by type and search query); and analyze_audio_files to perform comprehensive mixing analysis across audio in a project, including level, frequency content, stereo imaging, and loudness metrics. When you ask Claude about a specific project, the server will locate the project with find_reaper_projects, parse it with parse_reaper_project, and then apply the analysis tools to provide actionable feedback. The tools are intended to help you identify clipping, overly loud sections, phase issues, and areas where plugins or routing could be improved. You can also query installed plugins (e.g., which VST3s or AU plugins you have) to understand your processing options before making changes.

Prerequisites:

• Python 3.8+ (recommended) and a working Python environment
• Access to Claude Desktop or your chosen MCP client
• Network access between the MCP client and the server if running on separate hosts

Installation steps:

1. Create and activate a Python virtual environment

python -m venv .venv
source .venv/bin/activate

2. Install the MCP server package (from the repository you provided or published on PyPI)

uv pip install .

3. Verify installation (example command, adjust if your setup uses a different entry point)

uv list  # or the appropriate command to list installed MCP servers

4. Configure the server (example config snippet to tailor to your environment)

• Set the Reaper project directory in your config or environment so find_reaper_projects can locate RPP files.
• Ensure the Python environment used by the MCP server can access Reaper project files.

5. Run the server (examples; use the configuration that matches your environment)

uv run reaper  # or the exact command your setup uses to start the MCP server

6. Connect your MCP client (e.g., Claude Desktop) and enable the reaper connector as described in the README.

Notes:

• The repository uses a Python-based MCP server approach via uvx; the exact package name in your environment should match your project setup. If you publish a different entry point, adjust the mcp_config accordingly.

Tips and common issues:

• Ensure Reaper project directory paths are accessible by the MCP server process and properly referenced in your configuration.
• If the plugin discovery tool (list_installed_fx) does not show recently installed plugins, open Reaper and perform a plugin cache rescan so Reaper updates its internal cache.
• For analyze_audio_files, you can filter by track name (e.g., track_filter = 'Vocals') to focus on specific elements of your mix.
• LUFS targets and thresholds provided in the documentation are guidelines; adjust thresholds to fit your streaming platform requirements or client preferences.
• If running behind a firewall or on a private network, ensure the MCP client can reach the MCP server’s host/port; use secure connections if transmitting project data.
• The server references project data structures in code (see src/reaper_mcp_server/reaper_dataclasses.py) for understanding how data is parsed and exposed to clients.