FenderStudioProMCP

FenderStudioProMCP

An MCP server that turns PreSonus Studio One into an AI-controllable DAW, enabling production workflows like editing, MIDI manipulation, mixing, effects, and export through an LLM.

Category
访问服务器

README

FenderStudioProMCP

Control PreSonus Studio One with an LLM — an MCP (Model Context Protocol) server that turns Studio One into an AI-controllable DAW instrument. Aimed at production workflows (editing, MIDI manipulation, mixing, effects, export), with a longer-term vision of AI-assisted live-set control.

Community project, work in progress. Not affiliated with or endorsed by PreSonus.


How it controls Studio One

Studio One has no public remote-control API, so FenderStudioProMCP drives it through four complementary, push-based paths — each proven to work on Studio One 7 (macOS):

Path What it does Needs
MIDI → command dispatch ⭐ A user control-surface (StudioOneMCP Pads) maps MIDI CCs/notes to Studio One commands and macros. The MCP sends MIDI on a virtual port → the command fires. Reliable, no Accessibility. Virtual port + surface bound to Receive From
MCU MIDI Faders, pan, mute, solo, transport via the built-in Mackie Control Universal surface Virtual port + MCU surface
Menu / keyboard automation Clicks Studio One menu items and sends shortcuts via macOS System Events; verifies actions via the Edit▸Undo label macOS Accessibility
Macro generation Writes .studioonemacro files for command sequences with no shortcut Nothing

⭐ The MIDI → command path is the backbone. It's confirmed working: the MCP sends a CC on the StudioOneMCP virtual port and Studio One fires the mapped command (e.g. Add Audio Track, verified via the Edit▸Undo label). Every Studio One command and every installed macro can be reached this way — see the catalog below.


Function catalog

Studio One's command surface is enumerated into docs/function-catalog.json:

  • 194 built-in commands (Edit, Event, Audio, Transport, Track, Song, File, View — top level)
  • 221 macros (the installed macro library — quantize, humanize, velocity, articulations, chords, tempo, add-EQ/compressor, export, …)
  • = 415 functions, mappable to MIDI (notes/CCs × 16 channels ≈ 2000 slots)

Mapping all 415 to MIDI + exposing them as named MCP tools is the current build — see Issues.


Setup

1. Install

git clone https://github.com/pajdarova/FenderStudioProMCP
cd FenderStudioProMCP
python -m venv .venv && ./.venv/bin/pip install -e ".[dev]"

2. Virtual MIDI port

FenderStudioProMCP opens a virtual port named StudioOneMCP (pinned to a fixed uniqueID so Studio One re-binds to it every launch). Just run the server — the port appears. (On Windows/Linux, use a loopback such as loopMIDI / snd-virmidi named StudioOneMCP.)

3. Install the control surface (MIDI → command)

Copy studio-one-devices/StudioOneMCPPads/ into your Studio One User Devices folder:

~/Library/Application Support/PreSonus/Studio One 7/User Devices/StudioOneMCPPads/

Then in Studio One ▸ Settings ▸ External Devices ▸ Add… pick StudioOneMCP ▸ StudioOneMCP Pads, and set Receive From = StudioOneMCP. Leave Send To empty (the surface is receive-only). Relaunch Studio One so it scans the device.

4. (Optional) MCU surface for faders/transport

External Devices ▸ Add ▸ Mackie Control Universal ▸ Receive From = StudioOneMCP.

5. (Optional) Accessibility for menu/keyboard automation

System Settings ▸ Privacy & Security ▸ Accessibility → enable your MCP host app.

6. Add to your MCP client (e.g. Claude Desktop)

{
  "mcpServers": {
    "studio-pro": { "command": "studio-pro-mcp", "args": ["--port-name", "StudioOneMCP"] }
  }
}

Available tools (current)

  • studio_one_run_command(name) ⭐ — fire any of 224 built-in Studio One commands by name (Edit, Track, Event, Audio, Song, Transport, Zoom, Show, Arranger, …). Case-insensitive, unique partial match accepted. e.g. "Duplicate", "Split Range", "Find Track", "Bounce Selection". Verified working end-to-end.
  • studio_one_list_commands(filter) — discover command names.
  • Transport: play · stop · record · rewind · fast-forward · toggle-loop · save · undo · redo
  • Mixer (MCU): set-fader · toggle-mute · toggle-solo · toggle-rec-arm · select-channel · set-pan
  • Automation: add tracks · new song · save-as · toggle mixer/browser/editor · zoom · split · …
  • Macros: generate .studioonemacro sequences

Note: the 224 mapped commands are Studio One built-ins (they dispatch from a control surface). User macros (139 in docs/macros-todo.json) do not dispatch from a surface — they need a keyboard/Command-Bar trigger, tracked in #3.


Architecture

LLM / MCP client
   │ JSON-RPC (stdio)
   ▼
MCP server (server.py) ── tool registration, arg validation
   │
   ├─ MidiBridge (midi_bridge.py) ─┐
   │   MCU + CC command dispatch   │  virtual port "StudioOneMCP"
   ├─ CoreMIDI pin (coremidi.py) ──┘  (pinned uniqueID)  ──► Studio One
   │                                       surfaces: StudioOneMCP Pads (commands), MCU (mixer)
   ├─ Automation (keystrokes.py)  ──► System Events (menus / shortcuts)
   └─ Macro writer (macro_writer.py) ──► ~/Documents/Studio One/Macros

Roadmap

Tracked in GitHub Issues. Highlights:

  • Expand the pad surface to the full MIDI → function map (all 415)
  • Named MCP tools to trigger any function ("quantize to 1/16", "add EQ + comp", "export stems")
  • Vision: AI DAW-controller for autonomous mixing — songs, faders, effects, transitions
  • Feedback/sensing (MCU meters, state model) for level-aware decisions

Development

pytest ; ruff check src tests ; mypy src

Credits

FenderStudioProMCP began as a fork of tiwadara/StudioOneMcp (MIT licensed), which established the MIDI/MCU control-surface approach to driving Studio One from an MCP server. This repository has since been substantially rewritten: Windows support, adaptation to the Fender Studio Pro rebrand, an MCP SDK 2.0 upgrade with dual stdio/HTTP transport, and a command catalog generated per-user from .keyscheme exports.

License

MIT — see LICENSE.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选