FastAPI-MCP is a zero-configuration tool for automatically exposing FastAPI endpoints as Model Context Protocol (MCP) tools. MCP is an open protocol that standardizes how applications provide context to LLMs.

What are the key features of FastAPI-MCP?

Key features include direct integration with FastAPI, zero configuration, automatic discovery of endpoints, preservation of schemas and documentation, and flexible deployment options.

How do I install FastAPI-MCP?

We recommend using `uv add fastapi-mcp` or alternatively, you can install with pip: `pip install fastapi-mcp`.

How do I use FastAPI-MCP with my FastAPI application?

Simply import `FastApiMCP` from the `fastapi_mcp` package, initialize it with your FastAPI app instance, and then mount it. See the Basic Usage example in the documentation.

What is the purpose of the `base_url` parameter?

The `base_url` tells the MCP server where to send API requests when tools are called. It is highly recommended to provide it explicitly, especially in deployed environments.

How does FastAPI-MCP name the MCP tools?

FastAPI-MCP uses the `operation_id` from your FastAPI routes as the MCP tool names. If you don't specify an `operation_id`, FastAPI auto-generates one, which can be cryptic.

Can I customize which endpoints are exposed as MCP tools?

Yes, you can control which FastAPI endpoints are exposed using Open API operation IDs or tags. You can include or exclude specific operations or tags.

Can I deploy the MCP server separately from the original FastAPI app?

Yes, you can create an MCP server from one FastAPI app and mount it to a different app. This allows for greater flexibility and scalability.

What do I do if I add endpoints to my FastAPI app after creating the MCP server?

You need to refresh the MCP server to include the new endpoints by calling the `setup_server()` method on the `FastApiMCP` instance.

How do I connect to the MCP Server using SSE?

In applications like Cursor, navigate to Settings -> MCP and use the URL of your MCP server endpoint (e.g., `http://localhost:8000/mcp`) as the sse URL.

My MCP client does not support SSE, what can I do?

You can use [mcp-proxy stdio](https://github.com/sparfenyuk/mcp-proxy) to expose MCP server to client which does not support SSE. Instructions are in the documentation.

Where can I find examples of how to use FastAPI-MCP?

See the [examples](examples) directory in the FastAPI-MCP repository for complete examples.

Where can I get help and connect with other FastAPI-MCP users?

Join the [MCParty Slack community](https://join.slack.com/t/themcparty/shared_invite/zt-30yxr1zdi-2FG~XjBA0xIgYSYuKe7~Xg) to connect with other MCP enthusiasts, ask questions, and share your experiences with FastAPI-MCP.

What are the requirements for using FastAPI-MCP?

FastAPI-MCP requires Python 3.10+ (Recommended 3.12) and uv.

FastAPI MCP Server – Overview

Unleashing the Power of FastAPI with MCP: A Deep Dive into Seamless AI Integration

In the rapidly evolving landscape of artificial intelligence, the ability to connect Language Learning Models (LLMs) with real-world data and tools is paramount. The Model Context Protocol (MCP) emerges as a pivotal solution, standardizing how applications provide context to these AI models. This is where FastAPI-MCP steps in, offering a zero-configuration approach to expose FastAPI endpoints as MCP tools. This overview delves into the intricacies of FastAPI-MCP, exploring its features, use cases, and integration with platforms like UBOS to enhance AI agent development.

Understanding the Core: FastAPI and MCP

Before diving into the specifics of FastAPI-MCP, it’s essential to understand the foundational technologies at play:

FastAPI: A modern, high-performance Python web framework for building APIs. It’s known for its speed, ease of use, and automatic data validation.
Model Context Protocol (MCP): An open protocol that standardizes how applications provide context to LLMs. It acts as a bridge, enabling AI models to access and interact with external data sources and tools.

FastAPI-MCP elegantly bridges these two technologies, allowing developers to seamlessly integrate their FastAPI-based APIs with AI models through the MCP standard.

Key Features of FastAPI-MCP

FastAPI-MCP boasts a range of features designed to simplify the integration process and enhance the functionality of AI-powered applications:

Zero Configuration: The hallmark of FastAPI-MCP is its ease of use. With minimal configuration, it automatically exposes FastAPI endpoints as MCP tools.
Direct Integration: Mount an MCP server directly to your FastAPI application for seamless integration.
Automatic Discovery: FastAPI-MCP automatically discovers all FastAPI endpoints and converts them into MCP tools, saving developers time and effort.
Schema Preservation: It preserves the schemas of your request and response models, ensuring data integrity and consistency.
Documentation Preservation: The documentation of your endpoints, including Swagger definitions, is preserved, making it easier to understand and use the exposed MCP tools.
Flexible Deployment: Deploy your MCP server alongside your FastAPI application or separately, depending on your architectural needs.

Use Cases: Empowering AI Agents with Context

The capabilities of FastAPI-MCP unlock a wide range of use cases across various industries. Here are a few examples:

AI-Powered Customer Service: Integrate a FastAPI-based CRM system with an AI agent using FastAPI-MCP. The AI agent can then access customer data, order history, and support tickets to provide personalized and informed assistance.
Smart Home Automation: Expose smart home device controls through a FastAPI API and integrate it with an AI assistant using FastAPI-MCP. Users can then control their lights, thermostats, and appliances using natural language commands.
Financial Analysis: Connect a FastAPI-based financial data API with an AI model for investment analysis. The AI model can access real-time market data, historical trends, and financial news to generate insights and recommendations.
Content Creation: Allow AI models to access and manipulate content management systems, automatically generate articles, social media posts, or marketing copy using MCP tools exposed through FastAPI-MCP.
E-commerce Recommendations: Provide AI-powered recommendations based on product catalogs, user browsing history and other data provided through FastAPI endpoints exposed as MCP tools.

Advanced Usage: Tailoring FastAPI-MCP to Your Needs

While FastAPI-MCP is designed for ease of use, it also offers advanced customization options to cater to specific requirements:

Customizing Schema Descriptions: Control the level of detail included in the tool descriptions, including all possible response schemas and full JSON schema definitions.
Customizing Exposed Endpoints: Selectively expose endpoints based on Open API operation IDs or tags, allowing you to control which functionalities are available to AI models.
Separate Deployment: Deploy the MCP server separately from the original FastAPI application, providing greater flexibility and scalability.
Refreshing Endpoints: Refresh the MCP server to include new endpoints added after the initial creation, ensuring that the AI models have access to the latest functionalities.

Integrating with UBOS: A Powerful Synergy

UBOS is a full-stack AI Agent Development Platform focused on bringing AI Agents to every business department. It provides a comprehensive environment for orchestrating AI Agents, connecting them with enterprise data, and building custom AI Agents with LLM models and Multi-Agent Systems.

Integrating FastAPI-MCP with UBOS unlocks a powerful synergy. By exposing FastAPI-based APIs as MCP tools, you can seamlessly connect them to AI Agents orchestrated within the UBOS platform. This allows you to:

Orchestrate AI Agents: Use UBOS to define the workflow and interactions of AI Agents that leverage MCP tools exposed through FastAPI-MCP.
Connect to Enterprise Data: Connect AI Agents to your enterprise data sources through FastAPI APIs and MCP, enabling them to access and process relevant information.
Build Custom AI Agents: Create custom AI Agents with your own LLM models and integrate them with MCP tools for enhanced functionality.
Develop Multi-Agent Systems: Build complex Multi-Agent Systems where different AI Agents collaborate and interact with each other through MCP tools.

The UBOS Advantage: Enhanced AI Agent Development

UBOS complements FastAPI-MCP by providing a robust platform for managing and deploying AI Agents. Its features include:

Agent Orchestration: Visually design and manage complex AI Agent workflows.
Data Integration: Connect to various data sources, including databases, APIs, and cloud storage.
Model Management: Deploy and manage custom LLM models.
Monitoring and Analytics: Track the performance of AI Agents and identify areas for improvement.
Security and Compliance: Ensure the security and compliance of AI Agent deployments.

Getting Started with FastAPI-MCP and UBOS

To start leveraging the power of FastAPI-MCP and UBOS, follow these steps:

Install FastAPI-MCP: Use pip or uv to install FastAPI-MCP in your Python environment.
Integrate with FastAPI: Mount the MCP server to your FastAPI application with minimal configuration.
Configure UBOS: Set up your UBOS environment and connect it to your FastAPI-MCP server.
Orchestrate AI Agents: Design and deploy AI Agents within UBOS that leverage the exposed MCP tools.

Conclusion: A New Era of AI Integration

FastAPI-MCP represents a significant step forward in simplifying the integration of APIs with AI models. By providing a zero-configuration approach to exposing FastAPI endpoints as MCP tools, it empowers developers to build more intelligent and context-aware applications. When combined with the robust AI Agent Development Platform of UBOS, the possibilities are endless. This synergy unlocks a new era of AI integration, enabling businesses to leverage the power of AI to automate tasks, improve decision-making, and enhance customer experiences.

By embracing FastAPI-MCP and UBOS, you can unlock the full potential of AI and drive innovation across your organization.