erpnext_mcp_server
ERPNext management, file operations, read-only database access, and ERPNext API integration
claude mcp add --transport stdio manotluijiu-erpnext_mcp_server python -m your_app.mcp_server \ --env DEBUG="true"
How to use
ERPNext MCP Server provides a standardized development and production workflow for managing ERPNext-related MCP capabilities. It supports development with hot-reloading on port 8080 and a production-ready setup on port 8100, integrated with NGINX and supervisor. The server enables ERPNext management tasks, file operations, read-only database access, and ERPNext API integration, making it easier to build, test, and deploy MCP tools and resources within the Frappe/ERPNext ecosystem. You can interact with the MCP server through the development framework commands (mcp-config, mcp-dev, mcp-build, mcp-deploy) to configure entry points, ports, watched paths, and environment variables, as described in the project README. Tools and resources you implement under your app (mcp_server.py, mcp_dev.py, and the mcp tools/resources/prompts) will be exposed via the MCP interface, enabling scripted operations and integrated workflows within ERPNext. To run locally, start the development server using bench-based commands, then access the MCP UI/API exposed on http://localhost:8080 for development and on http://localhost:8100 or your site domain for production.
How to install
Prerequisites:
- Python 3.8+ and a functioning ERPNext/Frappe bench environment
- Access to a ERPNext site (e.g., bench site) and the ability to install apps
- Git and a working internet connection
Installation steps:
-
Create or install your ERPNext app that will host the MCP server code
-
Add the MCP development framework to your app (as per the project README):
In your app's hooks.py
commands = [ "your_app.mcp_dev.commands" ]
-
Implement the MCP server structure inside your app, including:
- your_app/mcp_server.py # Main server entry point
- your_app/mcp_dev.py # Development framework hookups
- your_app/mcp/ # Tools, resources, prompts implementations
-
Initialize and configure the development environment (as per README):
Initialize development environment
bench --site your-site.com mcp-config --app your_app_name
Configure development port (default: 8080)
bench --site your-site.com mcp-config --dev-port 8080
Configure production port (default: 8100)
bench --site your-site.com mcp-config --prod-port 8100
-
Run the development server to start on port 8080 with live reload:
bench --site your-site.com mcp-dev
-
For production deployment, build and deploy as described in the README:
bench --site your-site.com mcp-build bench --site your-site.com mcp-deploy --reload-nginx --restart-supervisor
Notes:
- Adjust environment variables via bench mcp-config --set-env or in your deployment environment.
- Ensure your entry point matches your app's MCP server implementation (e.g., your_app.mcp_server:server).
- NGINX and supervisor integration will route /mcp/ to your MCP server in production.
Additional notes
Tips and notes:
- Use bench mcp-config to set the entry point, watched paths, and environment variables; these settings will be used in both development and production setups.
- When deploying to production, ensure NGINX is configured to proxy /mcp/ correctly and that supervisor is managing the MCP server process.
- Common issues include port conflicts on 8080/8100; verify ports in both development and production configurations.
- Document your MCP server capabilities and provide examples of using tools/resources to facilitate onboarding for new developers.
- Keep your MCP tools modular and well-tested with unit and integration tests to ensure parity between development and production environments.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud