Django MCP Server is an implementation of the Model Context Protocol (MCP) extension for Django. This module allows MCP Clients and AI agents to interact with any Django application seamlessly.
How to install django-mcp-server?
gts360/django-mcp-server
$git clone https://github.com/gts360/django-mcp-serverInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install django-mcp-server by running `git clone https://github.com/gts360/django-mcp-server`, then use it for the current task and follow its documentation at https://github.com/gts360/django-mcp-server.
README.md
| 1 | # Django MCP Server |
| 2 | |
| 3 | [](https://pypi.org/project/django-mcp-server/) |
| 4 |  |
| 5 | [](https://djangopackages.org/packages/p/django-mcp-server/) |
| 6 |  |
| 7 | [](https://pypi.org/project/django-mcp-server/) |
| 8 | |
| 9 | **Django MCP Server** is an implementation of the **Model Context Protocol (MCP)** extension for Django. This module allows **MCP Clients** and **AI agents** to interact with **any Django application** seamlessly. |
| 10 | |
| 11 | 🚀 Django-Style declarative style tools to allow AI Agents and MCP clients tool to interact with Django.<br/> |
| 12 | 🚀 Expose Django models for AI Agents and MCP Tools to query in 2 lines of code in a safe way.<br/> |
| 13 | 🚀 Convert Django Rest Framework APIs to MCP tools with one annotation.<br/> |
| 14 | ✅ Working on both WSGI and ASGI without infrastructure change.<br/> |
| 15 | ✅ Validated as a Remote Integration with Claude AI.<br/> |
| 16 | 🤖 Any MCP Client or AI Agent supporting MCP , (Google Agent Developement Kit, Claude AI, Claude Desktop ...) can interact with your application. |
| 17 | |
| 18 | Many thanks 🙏 to [all the contributor community](https://github.com/omarbenhamid/django-mcp-server/graphs/contributors) |
| 19 | |
| 20 | Maintained ✨ with care by [Smart GTS software engineering](https://www.smart-gts.com/#contact). |
| 21 | |
| 22 | Licensed under the **MIT License**. |
| 23 | |
| 24 | --- |
| 25 | |
| 26 | ## Features |
| 27 | |
| 28 | - Expose Django models and logic as **MCP tools**. |
| 29 | - Serve an MCP endpoint inside your Django app. |
| 30 | - Easily integrate with AI agents, MCP Clients, or tools like Google ADK. |
| 31 | |
| 32 | --- |
| 33 | |
| 34 | ## Quick Start |
| 35 | |
| 36 | ### 1️⃣ Install |
| 37 | |
| 38 | ```bash |
| 39 | pip install django-mcp-server |
| 40 | ``` |
| 41 | |
| 42 | Or directly from GitHub: |
| 43 | |
| 44 | ```bash |
| 45 | pip install git+https://github.com/omarbenhamid/django-mcp-server.git |
| 46 | ``` |
| 47 | |
| 48 | --- |
| 49 | |
| 50 | ### 2️⃣ Configure Django |
| 51 | |
| 52 | ✅ Add `mcp_server` to your `INSTALLED_APPS`: |
| 53 | |
| 54 | ```python |
| 55 | INSTALLED_APPS = [ |
| 56 | # your apps... |
| 57 | 'mcp_server', |
| 58 | ] |
| 59 | ``` |
| 60 | |
| 61 | ✅ Add the **MCP endpoint** to your `urls.py`: |
| 62 | |
| 63 | ```python |
| 64 | from django.urls import path, include |
| 65 | |
| 66 | urlpatterns = [ |
| 67 | # your urls... |
| 68 | path("", include('mcp_server.urls')), |
| 69 | ] |
| 70 | ``` |
| 71 | |
| 72 | By default, the MCP endpoint will be available at `/mcp`. |
| 73 | |
| 74 | --- |
| 75 | |
| 76 | ### 3️⃣ Define MCP Tools |
| 77 | |
| 78 | In mcp.py create a subclass of `ModelQueryToolset` to give access to a model : |
| 79 | |
| 80 | ```python |
| 81 | from mcp_server import ModelQueryToolset |
| 82 | from .models import * |
| 83 | |
| 84 | |
| 85 | class BirdQueryTool(ModelQueryToolset): |
| 86 | model = Bird |
| 87 | |
| 88 | def get_queryset(self): |
| 89 | """self.request can be used to filter the queryset""" |
| 90 | return super().get_queryset().filter(location__isnull=False) |
| 91 | |
| 92 | class LocationTool(ModelQueryToolset): |
| 93 | model = Location |
| 94 | |
| 95 | class CityTool(ModelQueryToolset): |
| 96 | model = City |
| 97 | |
| 98 | ``` |
| 99 | |
| 100 | Or create a sub class of `MCPToolset` to publish generic methods (private _ methods are not published) |
| 101 | |
| 102 | Example: |
| 103 | ```python |
| 104 | from mcp_server import MCPToolset |
| 105 | from django.core.mail import send_mail |
| 106 | |
| 107 | class MyAITools(MCPToolset): |
| 108 | def add(self, a: int, b: int) -> list[dict]: |
| 109 | """A service to add two numbers together""" |
| 110 | return a+b |
| 111 | |
| 112 | def send_email(self, to_email: str, subject: str, body: str): |
| 113 | """ A tool to send emails""" |
| 114 | |
| 115 | send_mail( |
| 116 | subject=subject, |
| 117 | message=body, |
| 118 | from_email='your_email@example.com', |
| 119 | recipient_list=[to_email], |
| 120 | fail_silently=False, |
| 121 | ) |
| 122 | ``` |
| 123 | |
| 124 | --- |
| 125 | |
| 126 | ### Verify with MCP Inspect |
| 127 | |
| 128 | Use the management commande mcp_inspect to ensure your tools are correctly declared : |
| 129 | |
| 130 | ```bash |
| 131 | python manage.py mcp_inspect |
| 132 | ``` |
| 133 | |
| 134 | ### Use the MCP with any MCP Client |
| 135 | |
| 136 | The mcp tool is now published on your Django App at `/mcp` endpoint. |
| 137 | |
| 138 | **IMPORTANT** For production setup, on non-public data, consider enabling |
| 139 | authorization through : DJANGO_MCP_AUTHENTICATION_CLASSES |
| 140 | |
| 141 | ### Test with MCP Python SDK |
| 142 | |
| 143 | Y |
Related
6 picksType
- fastapi_mcpAutomatic MCP server generator for FastAPI applications - converts FastAPI endpoints to MCP tools for LLM integrationNov 2025
- mcp-server-odooA Model Context Protocol server for Odoo ERP systemsJul 2026
- odoo-mcp-proAI connector for Odoo ERP -- connect Claude, ChatGPT, and other AI tools to Odoo via MCP (Model Context Protocol)Jul 2026
- mcp-serverModel Context Protocol (MCP) server for AI-assisted development of CAP applications.Jul 2026
- fliA Python wrapper for Google Flights APIJul 2026
- meilisearch-mcpMCP server for MeilisearchJun 2026
