nt_helper

A cross-platform Flutter application designed for editing presets on the Expert Sleepers Disting NT module. It provides an intuitive interface for managing algorithms, parameters, and mappings on your Disting NT device.

Core Features

• Comprehensive Preset Management: Easily load, save, and create new presets.
• Detailed Algorithm Editing: Access and modify all parameters for each algorithm in your preset. Some algorithms feature custom UI views for a more specialized editing experience.
• Advanced Parameter Mapping: Configure CV, MIDI (including a CC detection helper), and I2C mappings for dynamic control over your sound.
• Visual Routing Analysis: Understand the signal flow within your presets with a clear graphical representation.
• Performance Mode: View and interact with all your mapped parameters on a single screen, with real-time updates from MIDI/I2C (CV changes provide offsets and are not currently reflected in real-time on sliders).
• Specialized Editors: Benefit from dedicated UI components for certain parameter types, such as an intuitive BPM editor for tempo-related parameters.
• MCP Server: Includes a built-in MCP (Model Context Protocol) server (on desktop platforms) for integration with external tools and AI-driven workflows. See the Model Context Protocol website for more details on MCP.
• Offline Data Management: Synchronize and manage algorithm metadata for offline use or when the Disting NT is not connected. Features incremental sync and improved algorithm rescan UX.
• Drag & Drop Installation: Install preset packages by simply dragging and dropping them onto the load dialog (desktop platforms).
• Cross-Platform: Runs on Windows, macOS, Linux, iOS, and Android.

Getting Started

1. Download the latest release for your platform from the GitHub Releases page (or TestFlight for iOS).
2. Install the application.
3. Connect your Disting NT to your device via USB.
4. Launch nt_helper, select the MIDI input/output ports for your Disting NT, and connect.

For detailed instructions, troubleshooting, and a full feature list, please refer to the project website documentation.

Recent Updates

• v1.39.0+77: Added incremental sync and improved algorithm rescan UX
• Drag & Drop Support: Preset packages can now be installed via drag-and-drop on desktop platforms
• MCP Enhancements: Simplified server implementation, improved connection stability, and added CPU usage monitoring
• Algorithm Matching: Added fuzzy category matching for better algorithm discovery
• Routing Improvements: Enhanced real-time routing data queries from hardware

Contributing

Contributions, issues, and feature requests are welcome! Please feel free to check the issues page.

License

This project is open source.

MCP Tool Reference

The nt_helper application exposes a comprehensive set of functions via its built-in MCP (Model Context Protocol) server, allowing for programmatic interaction. These tools enable AI assistants to control the application and interact with the Disting NT hardware.

Important Notes:

• All tool parameters are passed as a JSON object in the request.
• All tools return a JSON string. Successful operations return the requested data, while errors include an "error": "message" field.
• Parameter references use parameter_number (0-based index) from get_current_preset or parameter_name for unique parameter names.
• Values are automatically scaled using display values (not raw internal values).
• Some tools accept empty JSON object {} when no parameters are required.

Algorithm Metadata Tools

get_algorithm_details - Retrieves full metadata for a specific algorithm

• Parameters: algorithm_guid (string) OR algorithm_name (string), expand_features (bool, optional)
• Supports fuzzy matching ≥70% similarity for algorithm names
• Returns: Complete AlgorithmMetadata object with parameters, categories, and description

list_algorithms - Lists available algorithms with optional filtering

• Parameters: category (string, optional), query (string, optional)
• Supports fuzzy category matching and text search across names/descriptions
• Returns: Array of algorithm summaries with name, GUID, and first sentence of description

get_routing - Retrieves current routing state and signal flow

• Parameters: None
• Actively refreshes routing data from hardware before returning
• Returns: Bus usage information showing inputs/outputs for each algorithm slot

Core Disting NT Tools

get_current_preset - Gets complete preset state including all slots and parameters

• Parameters: None
• Returns: Preset name, slot configurations, algorithm details, and parameter info with live values
• Essential starting point for understanding current device state

add_algorithm - Adds algorithm to first available slot

• Parameters: algorithm_guid (string) OR algorithm_name (string)
• Supports exact and fuzzy name matching
• Returns: Success confirmation with slot placement

remove_algorithm - Clears algorithm from specified slot

• Parameters: slot_index (int, required)
• Returns: Success confirmation

set_parameter_value - Sets parameter value using display values

• Parameters: slot_index (int), parameter_number (int) OR parameter_name (string), value (number)
• Automatically handles scaling based on parameter metadata
• Returns: Success confirmation with parameter details

get_parameter_value - Gets current parameter value

• Parameters: slot_index (int), parameter_number (int)
• Returns: Scaled parameter value with metadata

Preset Management Tools

set_preset_name / get_preset_name - Manage preset names

• Set parameters: name (string, required)
• Get parameters: None
• Returns: Success confirmation or current preset name

set_slot_name / get_slot_name - Manage custom slot names

• Set parameters: slot_index (int), name (string)
• Get parameters: slot_index (int)
• Returns: Success confirmation or current slot name

new_preset - Creates new empty preset

• Parameters: None
• Clears all slots and resets to default state

save_preset - Saves current preset to device

• Parameters: None
• Persists all changes to device memory

Algorithm Movement Tools

move_algorithm_up / move_algorithm_down - Move algorithms one position

• Parameters: slot_index (int, required)
• Changes processing order (slot 0 processes first)
• Returns: Success confirmation

move_algorithm - Move algorithms multiple positions

• Parameters: slot_index (int), direction (string: "up"/"down"), steps (int, optional, default: 1)
• Performs multiple move operations in sequence
• Returns: Success confirmation with final position

Batch Operation Tools

set_multiple_parameters - Set multiple parameters in one operation

• Parameters: slot_index (int), parameters (array of objects with parameter_number/parameter_name and value)
• Efficient for configuring multiple parameters simultaneously
• Returns: Results array with success/failure status for each parameter

get_multiple_parameters - Get multiple parameter values efficiently

• Parameters: slot_index (int), parameter_numbers (array of integers)
• Returns: Array of parameter values with metadata

build_preset_from_json - Build complete preset from structured data

• Parameters: preset_data (object with preset_name and slots array), clear_existing (bool, optional, default: true)
• Supports complex preset creation with algorithms and parameter configurations
• Returns: Detailed build results with success/failure status per slot

Utility Tools

get_module_screenshot - Captures current device display

• Parameters: None
• Returns: Base64-encoded JPEG image of module screen
• Useful for visual confirmation of device state

get_cpu_usage - Monitors device performance

• Parameters: None
• Returns: CPU1/CPU2 percentages and per-slot usage breakdown

set_notes / get_notes - Manage preset notes

• Set parameters: text (string, max 7 lines × 31 characters)
• Get parameters: None
• Automatically creates/manages Notes algorithm in slot 0
• Returns: Notes content or success confirmation

find_algorithm_in_preset - Locate algorithms in current preset

• Parameters: algorithm_guid (string) OR algorithm_name (string)
• Searches all slots for specified algorithm
• Returns: Array of slot locations where algorithm is found

Diagnostic Tools

mcp_diagnostics - Check MCP server health and connection status

• Parameters: None
• Returns: Server status, active connections, and library version information

MCP Resources

The server also provides documentation resources accessible via MCP resource URLs:

• mcp://nt-helper/bus-mapping - Physical I/O to internal bus mapping reference
• mcp://nt-helper/usage-guide - Essential tools and best practices for small LLMs
• mcp://nt-helper/algorithm-categories - Complete list of 44+ algorithm categories
• mcp://nt-helper/preset-format - JSON structure documentation for build_preset_from_json
• mcp://nt-helper/routing-concepts - Signal flow and routing fundamentals

Best Practices

1. Start with get_current_preset to understand the current state
2. Use exact algorithm GUIDs when possible for reliable results
3. Check parameter ranges from get_current_preset before setting values
4. Save presets after making changes to persist them
5. Use batch operations for efficiency when setting multiple parameters
6. Monitor CPU usage when building complex presets
7. Use physical names (Input N, Output N, Aux N) when discussing routing