Skip to Content

LLM MCP

by
Odoo
v 16.0 Third Party 34
Download for v 16.0 Deploy on Odoo.sh
Availability
Odoo Online
Odoo.sh
On Premise
Odoo Apps Dependencies Discuss (mail)
Community Apps Dependencies
Lines of code 2975
Technical Name llm_mcp
LicenseLGPL-3
Websitehttps://github.com/apexive/odoo-llm
You bought this module and need support? Click here!
Availability
Odoo Online
Odoo.sh
On Premise
Odoo Apps Dependencies Discuss (mail)
Community Apps Dependencies
Lines of code 2975
Technical Name llm_mcp
LicenseLGPL-3
Websitehttps://github.com/apexive/odoo-llm

LLM MCP

Model Context Protocol Integration for Odoo AI

Connect your Odoo LLM chatbots with external tools and services using the standardized Model Context Protocol

Installation Guide View on GitHub

What is Model Context Protocol?

The Model Context Protocol (MCP) is a standardized way for AI systems to discover and interact with external tools and services.

This module extends Odoo's LLM capabilities by adding support for the Model Context Protocol, enabling AI assistants in Odoo to connect with and use tools provided by external MCP-compliant servers.

By supporting this protocol, your Odoo instance becomes part of a growing ecosystem of AI tools, allowing your LLM models to execute external functions seamlessly through a standardized JSON-RPC 2.0 based protocol.

MCP Architecture Diagram

Model Context Protocol Architecture

Key Features

Auto-Discovery

Automatically discover and register tools exposed by MCP servers without manual configuration.

Seamless Execution

Execute external tools through your LLM conversations without additional user intervention.

Standardized Protocol

Full compliance with the JSON-RPC 2.0 based Model Context Protocol specification.

Standard I/O Transport

Connect to external processes using standard input/output pipes for secure communication.

Simple Management

Easy-to-use interface for managing MCP server connections and monitoring tool availability.

Integration with LLM Tools

Works with existing LLM Tool module for a consistent experience across all AI tools.

Use Cases

Code Execution

Connect to Python, JavaScript, or other code execution environments to let your LLM process data, perform calculations, or create visualizations based on user requests.

User: Can you analyze our sales data?
LLM: I'll analyze that for you using Python.
[Executes external Python tool via MCP]
Here's a breakdown of your sales trends...

Database Operations

Give your AI the ability to query external databases or data services, analyzing information beyond what's available in Odoo.

User: Find market data for our competitors
LLM: Let me check the market database.
[Calls external database tool via MCP]
Based on the market data, your competitors...

Custom AI Capabilities

Extend your LLM with specialized AI models for image recognition, document processing, or other domain-specific tasks.

User: What's in this product image?
LLM: Let me analyze that for you.
[Calls external image recognition via MCP]
The image shows a product with...

External API Integration

Connect to weather services, mapping tools, or any other external API through a standardized interface without additional development.

User: How's the weather at our delivery location?
LLM: Let me check that for you.
[Calls weather service via MCP]
The weather at the delivery location is...

Installation Guide

Prerequisites

  • Odoo v16.0 or newer
  • llm module installed
  • llm_tool module installed

Installation Steps

  1. Download the module

    Get the module from the Odoo App Store or GitHub repository.

  2. Upload to your Odoo instance

    Extract the module to your Odoo addons directory.

  3. Install the module

    Go to Apps menu, remove the apps filter, and search for "LLM MCP". Click Install.

  4. Configure MCP servers

    Navigate to LLM Configuration → MCP Servers to set up your connections.

Module Dependencies

- base
- mail
- llm
- llm_tool
Note: You need to have at least one MCP-compliant server available to use this module effectively. You can develop your own or use available open-source implementations.

Usage Guide

MCP Server Configuration

1. Configure an MCP Server

Start by setting up a connection to your MCP-compliant server:

  1. Navigate to LLM Configuration → MCP Servers
  2. Click Create to add a new server
  3. Fill in the server details:
    • Name: A descriptive name for the server
    • Transport Type: Select "Standard IO"
    • Command: The command to execute (e.g., python mcp_server.py)
    • Arguments: Any command line arguments required
  4. Click Save to create the server configuration
Starting the MCP Server

2. Start the Server and Discover Tools

Once you've configured the server, start it and discover available tools:

  1. Click the Start Server button to establish the connection
  2. The module will automatically discover and register tools exposed by the server
  3. View the discovered tools in the Tools tab
  4. You can click Refresh Tools anytime to update the tool list
Important: The external process must be MCP-compliant and respond to the initialization protocol. Non-compliant processes will fail to connect.
Using MCP Tools in Chat

3. Use MCP Tools in LLM Conversations

Once tools are registered, they can be used in LLM conversations:

  1. Navigate to any LLM conversation interface in Odoo
  2. The MCP tools will be available alongside other LLM tools
  3. When the LLM determines a tool should be used, it will:
    • Select the appropriate tool
    • Provide the necessary parameters
    • Execute the tool via the MCP server
    • Use the results in its response

The entire process is seamless from the user's perspective - they simply interact with the LLM assistant naturally, and the appropriate tools are invoked as needed.

Technical Details

Model Context Protocol Overview

The Model Context Protocol (MCP) is a JSON-RPC 2.0 based protocol that standardizes how AI systems discover and interact with external tools. Here's how it works:

Protocol Flow

  1. Initialization: The client (Odoo) sends an initialization request to the server
  2. Tool Discovery: The client requests a list of available tools
  3. Tool Registration: Tools are registered in Odoo's tool registry
  4. Tool Execution: When needed, the client sends tool execution requests
  5. Result Processing: Results are returned to the LLM for incorporation in responses
MCP Protocol Flow

Implementation Details

PipeManager

The core component that manages communication with external processes:

  • Creates and manages subprocess connections
  • Handles non-blocking I/O with external processes
  • Manages the JSON-RPC protocol communication
  • Implements connection pooling and error recovery

LLM Tool Integration

Extensions to the standard LLM Tool model:

  • Adds MCP as a new tool implementation type
  • Handles tool parameter validation
  • Manages tool execution via MCP servers
  • Processes and formats execution results

Developing MCP Servers

To create your own MCP-compliant server, you need to implement the following endpoints:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "initialize",
  "params": {
    "clientInfo": {
      "name": "odoo-llm-mcp",
      "version": "1.0.0"
    },
    "protocolVersion": "0.1.0",
    "capabilities": {
      "tools": {}
    }
  }
}

Server should respond with:

{
  "jsonrpc": "2.0",
  "id": 1,
  "result": {
    "protocolVersion": "0.1.0",
    "serverInfo": {
      "name": "example-mcp-server",
      "version": "1.0.0"
    },
    "capabilities": {
      "tools": {}
    }
  }
}
{
  "jsonrpc": "2.0",
  "id": 2,
  "method": "tools/list",
  "params": {}
}

Server should respond with:

{
  "jsonrpc": "2.0",
  "id": 2,
  "result": {
    "tools": [
      {
        "name": "calculator",
        "description": "Performs mathematical calculations",
        "inputSchema": {
          "type": "object",
          "properties": {
            "expression": {
              "type": "string",
              "description": "Mathematical expression to evaluate"
            }
          },
          "required": ["expression"]
        },
        "annotations": {
          "title": "Calculator",
          "idempotentHint": true
        }
      }
    ]
  }
}
{
  "jsonrpc": "2.0",
  "id": 3,
  "method": "tools/call",
  "params": {
    "name": "calculator",
    "arguments": {
      "expression": "2 + 2 * 3"
    }
  }
}

Server should respond with:

{
  "jsonrpc": "2.0",
  "id": 3,
  "result": {
    "isError": false,
    "content": [
      {
        "type": "text",
        "text": "8"
      }
    ]
  }
}

For complete MCP specification, please visit:

https://github.com/llm-tools/model-context-protocol

Frequently Asked Questions

MCP provides a standardized protocol for tool discovery and execution, allowing any compatible server to expose tools to Odoo's LLM capabilities. Unlike custom integrations, MCP follows a JSON-RPC 2.0 based specification that makes it easy to integrate with a wide variety of external services without custom development for each one.

The module is designed with security in mind, but you should carefully evaluate any external MCP servers you connect to. The Standard I/O transport provides process isolation, and tools are executed with the permissions of the Odoo server. Always verify the source and security of external MCP servers before connecting them to your production environment.

Yes. Once MCP tools are registered in Odoo, they become standard LLM tools that can be configured with access controls. You can use Odoo's permission system to restrict which users or groups can access specific tools.

The module includes error handling for server disconnections. If a server becomes unavailable, tool execution will fail gracefully with an appropriate error message. The module also includes automatic reconnection capabilities when servers become available again.

The MCP ecosystem is growing, and several open-source implementations are available. Check the official MCP repository for a list of compliant servers and tools. You can also create your own MCP server by implementing the protocol specification.

Module Information

  • Name: LLM MCP
  • Version: 16.0.1.0.0
  • Category: Technical
  • Author: Apexive Solutions LLC
  • Website: GitHub
  • License: LGPL-3
  • Dependencies: base, mail, llm, llm_tool

Ready to enhance your Odoo AI capabilities?

Connect your Odoo instance to the growing ecosystem of AI tools with Model Context Protocol

Download Now Installation Guide
LLM MCP

Model Context Protocol Integration for Odoo AI

© 2025 Apexive Solutions LLC. All rights reserved.

Resources

Please log in to comment on this module

  • The author can leave a single reply to each comment.
  • This section is meant to ask simple questions or leave a rating. Every report of a problem experienced while using the module should be addressed to the author directly (refer to the following point).
  • If you want to start a discussion with the author, please use the developer contact information. They can usually be found in the description.
Please choose a rating from 1 to 5 for this module.