Dolibarr MCP Server
Professional Model Context Protocol (MCP) server for comprehensive Dolibarr ERP integration with full CRUD operations and business intelligence capabilities.
🚀 Features
Complete ERP Management
- Customer/Third Party Management: Full CRUD operations for customers and suppliers
- Product Catalog: Comprehensive product management with inventory tracking
- Invoice Management: Create, update, and track invoices with line items
- Order Processing: Complete order lifecycle management
- Contact Management: Maintain detailed contact records
- User Administration: User account and permissions management
Professional Grade
- Asynchronous Operations: High-performance async/await architecture
- Error Handling: Comprehensive error handling with detailed logging
- Type Safety: Full type hints with Pydantic validation
- CLI Interface: Professional command-line tools for testing and management
- Raw API Access: Direct access to any Dolibarr API endpoint
MCP Integration
- Tool-based Architecture: Each operation exposed as an MCP tool
- Schema Validation: Proper input/output schema validation
- Response Formatting: Structured JSON responses
- Error Propagation: Meaningful error messages and status codes
📋 Prerequisites
- Python 3.8 or higher
- Dolibarr instance with API enabled
- Dolibarr API key with appropriate permissions
🛠️ Installation
From Source
# Clone the repository
git clone https://github.com/latinogino/dolibarr-mcp.git
cd dolibarr-mcp
# Install in development mode
pip install -e .
# Or install dependencies directly
pip install -r requirements.txt
Using pip (when published)
pip install dolibarr-mcp
⚙️ Configuration
Environment Variables
Create a .env file in your project directory:
# Copy the example file
cp .env.example .env
Edit the .env file with your Dolibarr configuration:
# Dolibarr API Configuration
DOLIBARR_URL=https://your-dolibarr-instance.com/api/index.php
DOLIBARR_API_KEY=your_dolibarr_api_key_here
# Logging Configuration
LOG_LEVEL=INFO
Dolibarr API Setup
-
Enable the API module in Dolibarr:
- Go to Home → Setup → Modules
- Enable "Web Services API REST (developer)"
-
Create an API key:
- Go to Home → Setup → API/Web services
- Create a new API key for your user
- Ensure the user has appropriate permissions
-
Test the API:
curl -X GET "https://your-dolibarr-instance.com/api/index.php/status" \ -H "DOLAPIKEY: your_api_key_here"
🚀 Usage
Command Line Interface
Test your connection:
dolibarr-mcp test
Test with specific credentials:
dolibarr-mcp test --url "https://your-instance.com/api/index.php" --api-key "your_key"
Start the MCP server:
dolibarr-mcp serve
As an MCP Server
The server exposes comprehensive tools for Dolibarr management:
System Tools
test_connection- Test API connectivityget_status- Get system status and version
Customer Management
get_customers- List customers/third partiesget_customer_by_id- Get specific customer detailscreate_customer- Create new customerupdate_customer- Update existing customerdelete_customer- Delete customer
Product Management
get_products- List productsget_product_by_id- Get specific product detailscreate_product- Create new productupdate_product- Update existing productdelete_product- Delete product
Invoice Management
get_invoices- List invoicesget_invoice_by_id- Get specific invoice detailscreate_invoice- Create new invoice with line itemsupdate_invoice- Update existing invoicedelete_invoice- Delete invoice
Order Management
get_orders- List ordersget_order_by_id- Get specific order detailscreate_order- Create new orderupdate_order- Update existing orderdelete_order- Delete order
Contact Management
get_contacts- List contactsget_contact_by_id- Get specific contact detailscreate_contact- Create new contactupdate_contact- Update existing contactdelete_contact- Delete contact
User Management
get_users- List usersget_user_by_id- Get specific user detailscreate_user- Create new userupdate_user- Update existing userdelete_user- Delete user
Raw API Access
dolibarr_raw_api- Make direct API calls to any endpoint
Programmatic Usage
import asyncio
from dolibarr_mcp import Config, DolibarrClient
async def main():
config = Config.from_env()
async with DolibarrClient(config) as client:
# Get customers
customers = await client.get_customers(limit=10)
print(f"Found {len(customers)} customers")
# Create a new customer
new_customer = await client.create_customer(
name="Test Company",
email="test@company.com",
phone="+1-555-0123"
)
print(f"Created customer: {new_customer}")
# Get products
products = await client.get_products(limit=5)
print(f"Found {len(products)} products")
asyncio.run(main())
🏗️ Architecture
Project Structure
dolibarr-mcp/
├── src/dolibarr_mcp/
│ ├── __init__.py # Package exports
│ ├── config.py # Configuration management
│ ├── cli.py # Command line interface
│ ├── dolibarr_client.py # API client implementation
│ └── dolibarr_mcp_server.py # MCP server implementation
├── tests/ # Test suite
├── api/ # API documentation
├── pyproject.toml # Project configuration
├── requirements.txt # Dependencies
└── README.md # This file
API Client Architecture
The DolibarrClient provides:
- Async Context Manager: Proper session management
- Error Handling: Custom
DolibarrAPIErrorexceptions - Request/Response Logging: Detailed debugging information
- Type Safety: Full type hints for all methods
- Flexible Configuration: Environment-based configuration
MCP Server Architecture
The MCP server follows the official MCP specification:
- Tool Registration: Each operation is a registered tool
- Schema Validation: Input/output validation using JSON schemas
- Error Propagation: Meaningful error responses
- Async Operation: Non-blocking request handling
🧪 Testing
Run the test suite:
# Install test dependencies
pip install pytest pytest-asyncio pytest-cov
# Run tests
pytest
# Run with coverage
pytest --cov=src/dolibarr_mcp
Test specific functionality:
# Test API connection
dolibarr-mcp test
# Test with verbose output
dolibarr-mcp test --verbose
📝 API Reference
Configuration
The Config class manages all configuration:
from dolibarr_mcp import Config
# From environment variables
config = Config.from_env()
# Manual configuration
config = Config(
dolibarr_url="https://your-instance.com/api/index.php",
api_key="your_api_key",
log_level="DEBUG"
)
Client Usage
from dolibarr_mcp import DolibarrClient, Config
config = Config.from_env()
async with DolibarrClient(config) as client:
# All operations here
pass
Error Handling
from dolibarr_mcp import DolibarrAPIError
try:
result = await client.get_customers()
except DolibarrAPIError as e:
print(f"API Error: {e.message}")
print(f"Status Code: {e.status_code}")
print(f"Response Data: {e.response_data}")
🐳 Docker Support (Coming Soon)
Docker containerization is planned for the second phase of development:
# Build image
docker build -t dolibarr-mcp .
# Run container
docker run -p 8080:8080 --env-file .env dolibarr-mcp
🤝 Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
Development Setup
# Clone your fork
git clone https://github.com/yourusername/dolibarr-mcp.git
cd dolibarr-mcp
# Install in development mode
pip install -e ".[dev]"
# Install pre-commit hooks
pre-commit install
# Run tests
pytest
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
🙏 Acknowledgments
- Dolibarr - The amazing open-source ERP/CRM
- Model Context Protocol - The protocol specification
- Anthropic - For the MCP standard
📞 Support
🗺️ Roadmap
Phase 1 (Current) ✅
- Core MCP server implementation
- Full CRUD operations for main entities
- Professional error handling
- CLI interface
- Comprehensive documentation
Phase 2 (Next)
- Docker containerization
- Advanced filtering and search
- Webhook support
- Performance optimization
- Extended API coverage
Phase 3 (Future)
- Web UI for management
- Multi-instance support
- Caching layer
- Metrics and monitoring
- Plugin system
Built with ❤️ for the Dolibarr community