TO

tbottlik/odoo-mcp-server-streamable-http

开发工具

本页由 AI 翻译,技术细节请以原文为准。 查看原文

0 stars 13 forks 质量 67 趋势 0GPL-3.0

MCP server to manage Odoo account with AI

概览

A hybrid Model Context Protocol (MCP) server that enables AI assistants to interact with Odoo ERP systems. Supports both (for web AI agents) and (for local AI assistants) modes. : ~/Library/Application Support/Claude/claude_desktop_config.json : %APPDATA%/Claude/claude_desktop_config.json This implementation allows you to expose your local MCP server to AI agents like Flowhunt, enabling remote automation of your Odoo business processes. - : AI agents can connect to your MCP server from anywhere - : Enable Flowhunt and other AI platforms to automate your Odoo workflows - : ngrok provides secure HTTPS tunnels with authentication - : No need to deploy to cloud servers 1. Go to ngrok.com and create a free account 2. Get your authtoken from the dashboard 3. Configure ngrok: (e.g., https://abc123.ngrok-free.app) - this is your public MCP server endpoint. 1. - Go to Flowhunt Dashboard - Click "Create New Flow" - Give your flow a name (e.g., "Odoo Business Automation") 2.

README

以下为原文 README 节选,完整内容请查看原文。

Odoo MCP Server (Hybrid)

A hybrid Model Context Protocol (MCP) server that enables AI assistants to interact with Odoo ERP systems. Supports both HTTP streaming (for web AI agents) and stdio (for local AI assistants) modes.

📺 Demo

stdio Mode Demo (Claude Desktop)

HTTP Mode Demo (AI Agents & Remote Access)

Example Output

🚀 Quick Start

HTTP Mode (AI Agents & Remote Access)

# Clone the repository
git clone https://github.com/vzeman/odoo-mcp-server.git
cd odoo-mcp-server

# Copy example and configure
cp docker-compose.example.yml docker-compose.yml
# Edit docker-compose.yml with your Odoo credentials

# Start with Docker Compose
docker-compose up -d

# Access the server
curl http://localhost:8000/health

stdio Mode (Claude Desktop)

# Install
pip install odoo-mcp-server

Option 2: Install from source

# Clone the repository
git clone https://github.com/vzeman/odoo-mcp-server.git
cd odoo-mcp-server

# Install in development mode
pip install -e .

Configure Claude Desktop

Add to your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "odoo": {
      "command": "python",
      "args": ["-m", "mcp_server_odoo"],
      "env": {
        "ODOO_URL": "https://your-instance.odoo.com",
        "ODOO_DB": "your-database",
        "ODOO_USERNAME": "[email protected]",
        "ODOO_API_KEY": "your-api-key"
      }
    }
  }
}

Local Development

# Clone the repository
git clone https://github.com/vzeman/odoo-mcp-server.git
cd odoo-mcp-server

# Install dependencies
pip install -e .

# Set environment variables
export ODOO_URL="https://your-instance.odoo.com"
export ODOO_DB="your-database"
export ODOO_USERNAME="[email protected]"
export ODOO_API_KEY="your-api-key"

# Run the HTTP server
python -m mcp_server_odoo.http_server

# Run the stdio server
python -m mcp_server_odoo

🌐 Remote Access with ngrok (AI Agent Integration)

This implementation allows you to expose your local MCP server to AI agents like Flowhunt, enabling remote automation of your Odoo business processes.

Why Use ngrok?

  • Remote AI Access: AI agents can connect to your MCP server from anywhere
  • Business Automation: Enable Flowhunt and other AI platforms to automate your Odoo workflows
  • Secure Tunneling: ngrok provides secure HTTPS tunnels with authentication
  • No Infrastructure: No need to deploy to cloud servers

Setup Instructions

1. Install ngrok

Download and Install:

# macOS (using Homebrew)
brew install ngrok

# Or download directly from https://ngrok.com/download
在 GitHub 查看完整 README

安装

npx -y mcp-remote http://localhost:8000
新手?先看 mcpskill.org 教程 →

配置

{ "mcpServers": { "odoo": { "command": "python", "args": ["-m", "mcp_server_odoo"], "env": { "ODOO_URL": "https://your-instance.odoo.com", "ODOO_DB": "your-database", "ODOO_USERNAME": "[email protected]", "ODOO_API_KEY": "your-api-key" } } } }