MCP-Odoo

Model Context Protocol server for Odoo integration, allowing AI agents to access and manipulate Odoo data through a standardized interface.

Overview

MCP-Odoo provides a bridge between Odoo ERP systems and AI agents using the Model Context Protocol (MCP). This enables AI systems to:

• CRM Management: Complete lead/opportunity lifecycle management with Universidad ISEP customizations

• Partner Management: Comprehensive contact and customer/supplier management

• Accounting Analysis: View and analyze financial data including invoices and payments

• Academic Programs: University-specific program tracking and enrollment management

• Data Integration: Support for Mautic CRM and Google Analytics integration

• Multi-team Support: Sales team and workflow management

Related MCP server: Odoo MCP Server

Features

• 🔌 Easy integration with Odoo instances

• 🤖 Standard MCP interface for AI agent compatibility

• 📊 Rich accounting and CRM data access

• 🎓 University-specific academic program management (ISEP)

• 👥 Complete partner/contact lifecycle management

• 📈 CRM analytics and dashboard statistics

• 🔄 Lead to opportunity conversion workflows

• 🏢 Multi-team sales management

• 🔒 Secure authentication with Odoo

Installation

# Clone the repository
git clone https://github.com/yourtechtribe/model-context-protocol-mcp-odoo.git
cd model-context-protocol-mcp-odoo

# Install dependencies
pip install -r requirements.txt

Configuration

Create a .env file in the project root with the following variables:

ODOO_URL=https://your-odoo-instance.com
ODOO_DB=your_database
ODOO_USERNAME=your_username
ODOO_PASSWORD=your_password
HOST=0.0.0.0
PORT=8080

Usage

Start the MCP server:

# Using the SSE transport (default)
python -m mcp_odoo_public

# Using stdio for local agent integration
python -m mcp_odoo_public --transport stdio

Documentation

Comprehensive documentation is available in the docs/ directory:

• Documentation Home - Start here for an overview of all documentation

• Implementation Guide - Detailed architecture and implementation details

• Accounting Functionality - In-depth guide to accounting features

• CRM Functionality - Complete guide to CRM and partner management

• Troubleshooting - Solutions for common issues

• Usage Examples - Practical examples to get started

Available Tools

CRM Management

• list_leads - List and filter leads/opportunities with Universidad ISEP custom fields

• get_lead_details - Get comprehensive lead information

• create_lead - Create new leads with academic program tracking

• update_lead - Update lead information and progress

• convert_lead_to_opportunity - Convert leads to opportunities

• get_crm_dashboard_stats - Get performance metrics and analytics

Partner Management

• list_partners - List and filter contacts/companies

• get_partner_details - Get detailed partner information

• create_partner - Create new partners/contacts

• update_partner - Update partner information

Academic Programs (ISEP Specific)

• get_academic_programs - List available academic programs

• list_crm_stages - Get CRM workflow stages

• list_crm_teams - Get sales teams

• get_lead_activities - Get lead-related activities

Accounting (Existing)

• list_vendor_bills - List supplier invoices

• list_customer_invoices - List customer invoices

• list_payments - List payments with filtering

• reconcile_invoices_and_payments - Generate reconciliation reports

Development

Project Structure

• mcp_odoo_public/: Main package

  • odoo/: Odoo client and related modules

  • resources/: MCP resources definitions (tools and schemas)

  • server.py: MCP server implementation

  • config.py: Configuration management

  • mcp_instance.py: FastMCP instance definition

Adding New Resources

Resources define the capabilities exposed to AI agents through MCP. To add a new resource:

1. Create a new file in the resources/ directory

2. Define your resource using the @mcp.tool() decorator

3. Import your resource in resources/__init__.py

For detailed instructions, see the Implementation Guide.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Author

Albert Gil López

• Email: albert.gil@yourtechtribe.com

• LinkedIn: https://www.linkedin.com/in/albertgilopez/

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.