mcpField
M
Modbus MCP Server

Modbus MCP Server

Local
github.com/kukapayOther25Updated 76w ago

About Modbus MCP Server

A Modbus MCP server that standardizes and adds context to industrial Modbus data, connecting AI agents to industrial IoT systems. It is aimed at industrial automation use and supports TCP, UDP and serial connections.

Key features

  1. 01Read and write holding registers and coils
  2. 02Read input registers
  3. 03Read multiple holding registers in one call
  4. 04Supports Modbus over TCP, UDP or serial
  5. 05Connection settings configured through environment variables

Use cases

  1. 01Letting an AI agent read register values from industrial devices
  2. 02Switching coil states through natural language requests
  3. 03Analyzing register values with a customizable prompt

MCP support

The server exposes Modbus read and write operations to MCP clients as tools such as read_register and write_coil, and provides the analyze_register prompt.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Modbus MCP Server

An MCP server that standardizes and contextualizes Modbus data, enabling seamless integration of AI agents with industrial IoT systems.

GitHub License Python Version Status

Features

  • Modbus Tools:
    • Read/write holding registers (read_register, write_register).
    • Read/write coils (read_coils, write_coil).
    • Read input registers (read_input_registers).
    • Read multiple holding registers (read_multiple_holding_registers).
  • Prompt: Analyze Modbus register values with a customizable prompt (analyze_register).
  • Flexible Connections: Supports Modbus over TCP, UDP, or serial, configured via environment variables.

Requirements

  • Python: 3.10
  • uv for dependency and virtual environment management.

Installation

  1. Install uv:

    curl -LsSf https://astral.sh/uv/install.sh | sh
    
  2. Clone the Repository:

    git clone https://github.com/kukapay/modbus-mcp.git
    cd modbus-mcp
    
  3. Install Dependencies:

    uv sync
    

Configuration

The server connects to a Modbus device using parameters specified via environment variables. Set these variables in a .env file or your shell environment.

Environment Variables
VariableDescriptionDefaultRequired
MODBUS_TYPEConnection type: tcp, udp, or serialtcpYes
MODBUS_HOSTHost address for TCP/UDP127.0.0.1For TCP/UDP
MODBUS_PORTPort for TCP/UDP502For TCP/UDP
MODBUS_DEFAULT_SLAVE_IDSlave ID1For TCP/UDP
MODBUS_SERIAL_PORTSerial port (e.g., /dev/ttyUSB0, COM1)/dev/ttyUSB0For serial
MODBUS_BAUDRATESerial baud rate9600For serial
MODBUS_PARITYSerial parity: N (none), E (even), O (odd)NFor serial
MODBUS_STOPBITSSerial stop bits1For serial
MODBUS_BYTESIZESerial byte size8For serial
MODBUS_TIMEOUTSerial timeout (seconds)1For serial
Example .env File

For TCP:

MODBUS_TYPE=tcp
MODBUS_HOST=192.168.1.100
MODBUS_PORT=502
MODBUS_SLAVE_ID=1

For Serial:

MODBUS_TYPE=serial
MODBUS_SERIAL_PORT=/dev/ttyUSB0
MODBUS_BAUDRATE=9600
MODBUS_PARITY=N
MODBUS_STOPBITS=1
MODBUS_BYTESIZE=8
MODBUS_TIMEOUT=1

Usage

Installing for Claude Desktop

The configuration file:

{
   "mcpServers": {
       "Modbus MCP Server": {
           "command": "uv",
           "args": [ "--directory", "/path/to/modbus-mcp", "run", "modbus-mcp" ],
           "env": { "MODBUS_TYPE": "tcp", "MODBUS_HOST": "127.0.0.1", "MODBUS_PORT": 502 },
       }
   }
}
Using Tools

Note: Natural language support depends on the client’s ability to parse and map prompts to tools. The MCP Inspector requires structured JSON, but the examples below show how conversational inputs translate.

  1. Read a Holding Register:

    • Prompt:
      Please read the value of Modbus holding register 0.
      
    • MCP Inspector JSON:
      {
        "tool": "read_register",
        "parameters": {"address": 0, "slave_id": 1}
      }
      
    • Expected Output: Value: <register_value>
  2. Write to a Holding Register:

    • Prompt:
      Set Modbus holding register 10 to the value 100.
      
    • MCP Inspector JSON:
      {
        "tool": "write_register",
        "parameters": {"address": 10, "value": 100, "slave_id": 1}
      }
      
    • Expected Output: Successfully wrote 100 to register 10
  3. Read Coils:

    • Prompt:
      Check the status of the first 5 Modbus coils starting at address 0.
      
    • MCP Inspector JSON:
      {
        "tool": "read_coils",
        "parameters": {"address": 0, "count": 5, "slave_id": 1}
      }
      
    • Expected Output: Coils 0 to 4: [False, False, False, False, False]
  4. Write to a Coil:

    • Prompt:
      Turn on Modbus coil 5.
      
    • MCP Inspector JSON:
      {
        "tool": "write_coil",
        "parameters": {"address": 5, "value": true, "slave_id": 1}
      }
      
    • Expected Output: Successfully wrote True to coil 5
  5. Read Input Registers:

    • Prompt:
      Read the values of 3 Modbus input registers starting from address 2.
      
    • MCP Inspector JSON:
      {
        "tool": "read_input_registers",
        "parameters": {"address": 2, "count": 3, "slave_id": 1}
      }
      
    • Expected Output: Input Registers 2 to 4: [<value1>, <value2>, <value3>]
  6. Read Multiple Holding Registers:

    • Prompt:
      Get the values of Modbus holding registers 0 through 2.
      
    • MCP Inspector JSON:
      {
        "tool": "read_multiple_holding_registers",
        "parameters": {"address": 0, "count": 3, "slave_id": 1}
      }
      
    • Expected Output: Holding Registers 0 to 2: [<value1>, <value2>, <value3>]

License

This project is licensed under the MIT License. See LICENSE for details.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.