Rekordbox Performer
Enables safety-first live control of Rekordbox over a virtual MIDI port, with armed transitions, dry-run preview, and emergency stop.
README
Rekordbox Performer
Safety-first live Rekordbox control over a dedicated virtual MIDI port.
[!WARNING] This is an experimental prototype for supervised performance. Keep a physical controller available and test mappings with channel faders down.
Signal path
MCP client → transition scheduler → loopMIDI → Rekordbox MIDI Learn
The DDJ-FLX4 remains connected as the Hardware Unlock device and manual control surface. This server does not modify the Rekordbox database or process audio.
Requirements
- Windows with Rekordbox 7 in Performance mode
- Python 3.12 or newer
- A virtual MIDI port such as loopMIDI
- A Rekordbox-compatible controller for Hardware Unlock when required
Install
py -3.12 -m venv .venv
.\.venv\Scripts\python.exe -m pip install -e .
Setup
- Install and start loopMIDI.
- Create a port named
Codex Rekordbox Performerand enable loopMIDI autostart. - Open Rekordbox 7 in Performance mode with the FLX4 connected.
- Open Preferences → Controller → MIDI and select the virtual port.
- Import
mapping/Codex-Rekordbox-Performer.midi.csv. The fallbackmapping/rekordbox-midi-learn.csvcan be used for manual MIDI Learn. - Keep channel faders down and decks stopped during mapping.
- Register
rekordbox-performeras a stdio MCP server in your MCP client. - Call
list_midi_outputs,connect_midi, andcontrol_status. - Preview a transition, arm control, then perform it.
The dedicated mapping does not replace the DDJ-FLX4 factory mapping.
Safety model
- Live messages are rejected until control is armed.
- Arming expires automatically after at most 30 minutes.
- Transition plans default to dry-run preview.
- All scheduled events run against a local monotonic clock.
- Emergency stop cancels pending automation and disarms without moving controls.
- The physical FLX4 remains the manual override.
Do not move a physical FLX4 continuous control while automation is moving the same software control; mismatched hardware positions can cause value jumps.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。