Skip to main content
Glama
FL Studio Licheski MCP

FL Studio Licheski MCP

by Glicheski

FL Studio Licheski MCP

Private. Copyright (c) 2026 Guilherme Licheski. All rights reserved (see LICENSE).

MCP server that drives FL Studio as a full production rig from an AI assistant: tempo, channels, mixer, plugin parameters and presets, loading instruments from FL's browser, writing MIDI parts into patterns with sample-tight timing, and rendering seamless loops / one-shots from FL's own playback.

How it works

  • Bridge: the server writes cmd.json next to the controller script and sends note 127 on MIDI channel 16 through loopMIDI; the FL script runs the command and answers in res.json. FL only lets scripts write inside its REAL user data folder, so both sides resolve it (OneDrive first).

  • Notes: parts are played live once (each part on its own MIDI channel, routed to an FL channel), FL's score log is dumped into a fresh pattern and quantized (Snap_ThirdBeat for swing), so playback is sample-accurate.

  • Render: FL plays the pattern while its output is recorded (Windows WASAPI loopback, separate process); the loop period is measured by correlation and one period is kept with an 8 ms seam crossfade. One-shots are trimmed and faded.

  • Instruments: browser navigation by API; the final "Open in new channel" popup is confirmed with one Enter key, sent only after verifying that FL owns the foreground window.

Related MCP server: fl-bridge-mcp

Setup (Windows)

  1. FL Studio 20.7+, loopMIDI (one port), Python 3.10+, uv, ffmpeg on PATH.

  2. uv sync then uv run fl-licheski-install (copies fl_script/device_LicheskiMCP.py into FL's Hardware folder).

  3. FL: F10 > Input "loopMIDI Port" > Controller type FL Studio Licheski MCP > Enable. Output to loopMIDI OFF, "Send master sync" OFF, External clock sync OFF (otherwise MIDI loops back into FL).

  4. Claude Code: claude mcp add fl-licheski --scope user -- uv run --directory <this folder> fl-licheski-mcp

Tools

fl_ping, fl_api_dump, fl_tempo, fl_transport, fl_channels, fl_channel, fl_steps, fl_mixer, fl_plugin_params, fl_plugin_set, fl_plugin_preset, fl_browser, fl_perform, fl_write_parts, fl_render_pattern, fl_render_loop, fl_exec, fl_panic. Python helpers: loader.load(), loader.pick_preset(), performer.render_once().

Usage examples are kept outside this repository.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      C
      quality
      C
      maintenance
      Gives AI assistants full control of FL Studio through its scripting API, enabling project editing, mixing, piano roll composition, MIDI export, and plugin automation without touching the mouse or keyboard.
      57
      2
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables an AI agent to compose and mix directly inside FL Studio: writing and clearing notes, setting tempo, levels and plugin parameters, inspecting what FL currently holds, checking which channels sound, and rendering the arrangement to WAV.
      3
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI assistants to control a running FL Studio project — transport, mixer routing and levels, channel rack, plugins, patterns, step sequencer, and piano roll — through a virtual MIDI trigger plus JSON command channel, with optional Windows UI automation and offline audio analysis of bounces. Also exposes raw access to the FL scripting API for undo history, markers, and device and project info.
      MIT