DDC/CI Control Bridge

NPM Version MIT licensed GitHub Release

Install MCP Server

Multi-protocol server for DDC/CI monitor control with MCP and MQTT support.

Features

DDC/CI Monitor Control: Read and write VCP codes to control monitor brightness, contrast, input source, and more
Model Context Protocol (MCP): Server with resources and tools for AI/agent integration
MQTT Client: Connect to any MQTT broker for IoT/automation platforms
- Home Assistant auto-discovery support
- Generic MQTT pub/sub for custom integrations
Cross-Platform: Works on Windows, Linux, and macOS
Dual Distribution: Available as standalone executable or npm package

Quick Start

Installation

Option 1: NPM Package (Global)

npm install -g ddc-ci-control-bridge
ddc-ci-bridge

Option 2: NPX (No Install)

npx ddc-ci-control-bridge

Option 3: Standalone Executable

Download the binary for your platform from GitHub Releases:

Windows: ddc-ci-bridge-win.exe
Linux: ddc-ci-bridge-linux
macOS: ddc-ci-bridge-macos

⚠️ Note: The executable requires DDC_API_KEY environment variable to be set, or it will exit immediately.

Windows:

# Set API key (required)
set DDC_API_KEY=your-secret-key

# Run the executable
ddc-ci-bridge-win.exe

# Or set multiple environment variables
set DDC_API_KEY=your-secret-key
set MQTT_ENABLED=true
set MQTT_HOST=192.168.1.100
ddc-ci-bridge-win.exe

Linux/macOS:

# Make executable (first time only)
chmod +x ddc-ci-bridge-linux  # or ddc-ci-bridge-macos

# Run with API key (required)
DDC_API_KEY=your-secret-key ./ddc-ci-bridge-linux

# Or with multiple environment variables
DDC_API_KEY=your-secret-key MQTT_ENABLED=true MQTT_HOST=192.168.1.100 ./ddc-ci-bridge-linux

Using with MCP Clients:

Configure your MCP client to use the executable path:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "C:\\path\\to\\ddc-ci-bridge-win.exe",
      "args": [],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

MCP Client Setup

Choose between Local (stdio) or Remote (HTTP) setup depending on your needs:

Local Setup: Run MCP server on the same machine as your monitors (recommended for most users)
Remote Setup: Run MCP server on a monitor-connected machine, access from other devices

🖥️ Local Setup (stdio - Same Machine)

Use this if your MCP client is on the same machine as your monitors.

Go to: Settings → Cursor Settings → MCP → Add new global MCP server

Install MCP Server

Or paste the following configuration into your ~/.cursor/mcp.json file:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false",
        "MQTT_HOST": "localhost",
        "MQTT_PORT": "1883"
      }
    }
  }
}

Optional MQTT: Set MQTT_ENABLED=true and configure MQTT_HOST to your MQTT broker IP for automation platform integration.

Add this to your VS Code MCP config file:

"mcp": {
  "servers": {
    "ddc-ci-bridge": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Open Claude Desktop developer settings and edit your claude_desktop_config.json file:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Run this command:

claude mcp add ddc-ci-bridge -- npx -y ddc-ci-control-bridge

Add this to your Windsurf MCP config file:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Add this to your Cline MCP configuration:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Add this to your Zed settings.json:

{
  "context_servers": {
    "DDC/CI Bridge": {
      "source": "custom",
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Add this to your Roo Code MCP configuration:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Option A: Using the UI

Click the hamburger menu
Select Settings
Navigate to Tools section
Click + Add MCP button
Enter command: npx -y ddc-ci-control-bridge
Name: DDC/CI Bridge
Click Add

Option B: Manual Configuration

Add to settings.json:

"augment.advanced": {
  "mcpServers": [
    {
      "name": "ddc-ci-bridge",
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  ]
}

Go to Settings → Tools → AI Assistant → Model Context Protocol (MCP)
Click + Add
Select As JSON from the dropdown
Add this configuration:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Click Apply to save changes

Navigate Settings → AI → Manage MCP servers
Click + Add button
Paste this configuration:

{
  "DDC/CI Bridge": {
    "command": "npx",
    "args": ["-y", "ddc-ci-control-bridge"],
    "env": {
      "FASTMCP_API_KEY": "your-api-key-here",
      "MQTT_ENABLED": "false"
    },
    "working_directory": null,
    "start_on_launch": true
  }
}

Click Save

Add to your Codex MCP configuration:

[mcp_servers.ddc-ci-bridge]
command = "npx"
args = ["-y", "ddc-ci-control-bridge"]

[mcp_servers.ddc-ci-bridge.env]
MQTT_ENABLED = "false"

Windows Users: Use full paths if you encounter timeout errors:

[mcp_servers.ddc-ci-bridge]
command = "cmd"
args = ["/c", "npx", "-y", "ddc-ci-control-bridge"]
startup_timeout_ms = 20_000

[mcp_servers.ddc-ci-bridge.env]
SystemRoot = "C:\\Windows"
FASTMCP_API_KEY = "your-api-key-here"
MQTT_ENABLED = "false"

Navigate to Program (right side) → Install → Edit mcp.json
Add this configuration:

{
  "mcpServers": {
    "DDC/CI Bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Click Save
Toggle the MCP server on/off from the right side panel

Open Qodo Gen chat panel in VSCode or IntelliJ
Click Connect more tools
Click + Add new MCP
Add this configuration:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Navigate Perplexity → Settings
Select Connectors
Click Add Connector
Select Advanced
Enter Server Name: DDC/CI Bridge
Paste this JSON:

{
  "command": "npx",
  "args": ["-y", "ddc-ci-control-bridge"],
  "env": {
    "FASTMCP_API_KEY": "your-api-key-here",
    "MQTT_ENABLED": "false"
  }
}

Click Save

Navigate Kiro → MCP Servers
Click + Add button
Paste this configuration:

{
  "mcpServers": {
    "DDC/CI Bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Click Save

Open Settings → Plugins
Enter this JSON:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Save and restart if needed

Go to Zencoder menu (...)
Select Agent tools
Click Add custom MCP
Add name and server configuration:

{
  "command": "npx",
  "args": ["-y", "ddc-ci-control-bridge"],
  "env": {
    "FASTMCP_API_KEY": "your-api-key-here",
    "MQTT_ENABLED": "false"
  }
}

Click Install

Add to your Amazon Q configuration:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "npx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Bun:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "bunx",
      "args": ["-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

Deno:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "deno",
      "args": [
        "run",
        "--allow-env",
        "--allow-net",
        "npm:ddc-ci-control-bridge"
      ],
      "env": {
        "MQTT_ENABLED": "false"
      }
    }
  }
}

On Windows, use this format (example with Cline):

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "command": "cmd",
      "args": ["/c", "npx", "-y", "ddc-ci-control-bridge"],
      "env": {
        "MQTT_ENABLED": "false"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

🌐 Remote Setup (HTTP - Different Machines)

Use this if you want to access monitor controls from a different machine on your network.

On the machine with monitors, create a .env file:

DDC_API_KEY=your-secure-random-api-key-here
MCP_TRANSPORT=sse
MCP_PORT=8000

# Optional: Enable MQTT for automation platforms
MQTT_ENABLED=false
MQTT_HOST=localhost
MQTT_PORT=1883

Start the server:

npm install -g ddc-ci-control-bridge
ddc-ci-bridge

Or using npx:

npx ddc-ci-control-bridge

The server will display:

🔌 MCP Server: http://192.168.1.100:8000
   Transport: sse
   API Key: your-api...

Note the IP address and port for client configuration.

Cursor:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "url": "http://192.168.1.100:8000",
      "headers": {
        "Authorization": "Bearer your-secure-random-api-key-here"
      }
    }
  }
}

VS Code:

"mcp": {
  "servers": {
    "ddc-ci-bridge": {
      "type": "http",
      "url": "http://192.168.1.100:8000",
      "headers": {
        "Authorization": "Bearer your-secure-random-api-key-here"
      }
    }
  }
}

Claude Desktop:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "url": "http://192.168.1.100:8000",
      "headers": {
        "Authorization": "Bearer your-secure-random-api-key-here"
      }
    }
  }
}

Windsurf:

{
  "mcpServers": {
    "ddc-ci-bridge": {
      "serverUrl": "http://192.168.1.100:8000",
      "headers": {
        "Authorization": "Bearer your-secure-random-api-key-here"
      }
    }
  }
}

Note: Replace 192.168.1.100 with the actual IP address of your monitor-connected machine, and use the same API key you configured in the server's .env file.

Configuration

Create a .env file:

cp .env.example .env

Edit .env and configure:

DDC_API_KEY=your-secret-api-key-here

# Enable MQTT for automation platforms (optional)
MQTT_ENABLED=false
MQTT_HOST=localhost
MQTT_PORT=1883

To enable MQTT: Set MQTT_ENABLED=true and change MQTT_HOST to your MQTT broker IP (e.g., Home Assistant IP).

Start the server:

npm run dev

MQTT Integration

This server acts as an MQTT client and can connect to any MQTT broker.

Supported Platforms

Home Assistant: Auto-discovery via MQTT Discovery Protocol → See Home Assistant Guide
openHAB: MQTT binding with manual thing configuration
Node-RED: Subscribe to topics for custom flows
Domoticz: MQTT integration
Generic MQTT: Any platform that supports MQTT pub/sub

MQTT Topics

ddc-monitor/{index}/brightness/set|get
ddc-monitor/{index}/contrast/set|get
ddc-monitor/{index}/vcp/{code}/set|get
ddc-monitor/{index}/power/set
ddc-monitor/{index}/state
ddc-monitor/bridge/status

MCP Resources

monitor://list - List all available monitors
monitor://{index}/capabilities - Full VCP code scan results
monitor://{index}/current-values - Current values of all supported codes
monitor://{index}/info - Monitor metadata

MCP Tools

get_vcp_code - Read a specific VCP code value
set_vcp_code - Write a value to a VCP code
get_brightness - Get monitor brightness (0-100)
set_brightness - Set monitor brightness (0-100)
refresh_monitors - Re-scan monitors and VCP capabilities

Development

Build

npm run build

Build Executables

npm run build:executables

This creates standalone executables in dist/bin/ for Windows, Linux, and macOS.

Run in Development

npm run dev

Requirements

DDC/CI Support: Your monitor must support DDC/CI (most modern monitors do)
DDC/CI Enabled: Enable DDC/CI in your monitor's OSD settings
Permissions: On Linux, you may need to add your user to the i2c group

Linux Setup

sudo usermod -a -G i2c $USER
sudo modprobe i2c-dev

Log out and log back in for group changes to take effect.

Environment Variables

See .env.example for all available configuration options.

Key variables:

DDC_API_KEY - API key for server authentication (required)
MCP_TRANSPORT - Transport type: sse or stdio (default: sse)
MCP_PORT - HTTP port for SSE transport (default: 8000)
MQTT_ENABLED - Enable MQTT client (default: false)
MQTT_HOST - MQTT broker IP/hostname (default: localhost)
MQTT_PORT - MQTT broker port (default: 1883)
MQTT_USERNAME - Auto-configured as ddc-mcp
MQTT_PASSWORD - Auto-configured from DDC_API_KEY

Architecture

┌─────────────────────────────────────────┐
│       DDC/CI Control Bridge             │
│                                         │
│  ┌─────────────────┐  ┌──────────────┐ │
│  │  MCP Server     │  │ MQTT Client  │ │
│  │  - Resources    │  │ - Pub/Sub    │ │
│  │  - Tools        │  │ - Discovery  │ │
│  └────────┬────────┘  └──────┬───────┘ │
│           │                  │         │
│           └────────┬─────────┘         │
│                    │                   │
│           ┌────────▼────────┐          │
│           │ DDC Controller  │          │
│           │ - VCP Scanning  │          │
│           │ - Read/Write    │          │
│           └────────┬────────┘          │
└────────────────────┼───────────────────┘
                     │
            ┌────────▼────────┐
            │ Physical Monitor│
            │   (DDC/CI)      │
            └─────────────────┘

External Clients:
┌──────────────┐  ┌──────────────┐  ┌──────────────┐
│ AI Agents    │  │ MQTT Broker  │  │ Automation   │
│ (via MCP)    │  │ (any broker) │  │ Platforms    │
└──────────────┘  └──────────────┘  └──────────────┘

License

MIT

Documentation

Home Assistant Integration Guide - Complete HA setup
Development Guide - Technical architecture and development details

Contributing

Contributions welcome! Please open an issue or PR.

Troubleshooting

No monitors found

Ensure DDC/CI is enabled in your monitor's OSD settings
On Linux, check if i2c devices are available: ls -l /dev/i2c-*
Try running with elevated permissions (not recommended for production)

MQTT connection failed

Verify MQTT broker is running
Check credentials match the server output
Ensure firewall allows MQTT port (1883)

Commands not working

Check server logs for errors
Verify monitor supports the VCP code
Some codes may be read-only
Try restarting both server and Home Assistant