FastAPI MCP Server + LangChain Client Example

by SDCalvo

268 downloads Not rated yet Apache-2.0 license

GitHub

About

Example project demonstrating how to expose FastAPI endpoints as Model Context Protocol (MCP) tools using `fastapi-mcp`. Includes a basic LangChain agent (`langchain_client.py`) that connects to the local FastAPI server via HTTP/SSE using `langchain-mcp-adapters` to discover and

Details

License: Apache-2.0 license

Explore

This section summarizes features and concepts from the fastapi-mcp documentation that we haven't implemented yet but are useful to know.

Setting up with Highlight

This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:

Download and install Highlight from highlightai.com/download
Navigate to the plugins tab and select "Add Custom Plugin"
Configure the plugin with the settings below

Plugin Name FastAPI MCP Server + LangChain Client Example

Command (node, npx, python, etc.)

Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
Enable "Start Automatically" if you want the plugin to start when Highlight launches

Setting up with Cursor

Open Cursor's settings
Navigate to the MCP section
Click "Add new Global MCP server"
Copy and paste the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

Setting up with Claude Desktop

Open Claude Desktop's settings
Go to the Integrations tab
Click "Add MCP Server"
Use the configuration below based on your operating system

Note: Make sure Claude Desktop is running before attempting to connect.

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

Setting up with Windsurf

Open Windsurf's preferences
Go to the Integrations section
Click "Add MCP Integration"
Use the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

Setting up with Cline

Open Cline's settings
Navigate to the Integrations section
Click "Add MCP Server"
Copy and paste the configuration below based on your operating system

Note: Make sure to restart Cline after adding the configuration.

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

Setting up with Goose

Open Goose's preferences
Go to the Integrations tab
Select "MCP Servers"
Click "Add New Server"
Use the configuration below based on your operating system

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

JSON

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

From the repository

Prerequisites

Before you begin, ensure you have the following installed: - Python: Version 3.10 or higher recommended. - uv: The Python package manager used in this project. (Installation Instructions) - Node.js and npm: Required for npx (used to run the optional MCP Inspector). You can download Node.js (which includes npm) from nodejs.org. - Git: For cloning the repository. - OpenAI API Key: Required for the LangChain client example. You need to set this in a .env file. This project demonstrates setting up a basic FastAPI application and exposing its endpoints as Model Context Protocol (MCP) tools using the fastapi-mcp library. It also includes a LangChain agent client that connects to and uses these tools, and covers configuring Cursor to connect as well.

Getting Started / How to Run

1. Clone the Repository: ``

bash
    git clone <your-repo-url>
    cd <repo-directory>


2.  Install

uv: If you don't have it, install the uv

 package manager (see Project Setup below for command).
3.  Set up Environment & Install Dependencies:

bash
    uv init # If pyproject.toml doesn't exist
    uv venv # Create virtual environment (.venv)
    # Install all project dependencies
    uv pip install fastapi "uvicorn[standard]" fastapi-mcp langchain-mcp-adapters langgraph langchain-openai python-dotenv


4.  Create

.env File: Create a file named .env

 in the project root directory and add your OpenAI API key:

dotenv
    OPENAI_API_KEY=your_openai_api_key_here


5.  Run the FastAPI MCP Server: Open a terminal and run:

bash
    uvicorn main:app --reload --port 8000


    Keep this terminal running.
    _(Alternatively, you can use the

Python: FastAPI MCP debug configuration defined in .vscode/launch.json

 within VS Code / Cursor to run the server with the debugger attached.)_
6.  Run the LangChain Client: Open a _second_ terminal and run:

bash
    uv run python langchain_client.py



    The client will connect to the server, discover tools, and run a query using the agent.

7.  (Optional) Test Server with MCP Inspector: Before running the LangChain client, or for more direct testing, you can use the official MCP Inspector tool:

    - Ensure the FastAPI server is running (Step 5).
    - Open another terminal and run:

npx @modelcontextprotocol/inspector

_(

npx

 comes with Node.js/npm. If this command fails, ensure Node.js is installed and accessible in your PATH.)_
    - In the inspector UI, connect to your server URL:

http://127.0.0.1:8000/mcp


    - Navigate to "Tools", click "List Tools" to see

read_root__get and greet_user_greet__name__get

.
    - Select a tool, fill parameters (e.g.,

name

 for greet_user), and click "Run Tool".

8.  (Optional) Test

greet_user with LangChain Client: The greet_user endpoint and the corresponding test query (query2) in langchain_client.py

 are currently active. Simply run the LangChain client (Step 6) and observe the second part of its execution where it should attempt to greet the user 'LangChain'.
    - _(If you want to disable this test, comment out the

@app.get("/greet/{name}") endpoint in main.py and the query2 section in langchain_client.py

)_
Project Setup


1.  Package Manager: We used

, a fast Python package installer and resolver written in Rust.
    - Installation (Windows PowerShell):

irm https://astral.sh/uv/install.ps1 | iex


    - Project Initialization:

uv init (Creates pyproject.toml

)
    - Virtual Environment:

uv venv (Creates and manages .venv

)
2.  Dependencies: Installed using

bash
    # Specific commands used during development (covered by the combined install in Getting Started):
    # uv pip install fastapi "uvicorn[standard]" fastapi-mcp
    # uv pip install langchain-mcp-adapters langgraph langchain-openai python-dotenv


    This installs FastAPI, the Uvicorn ASGI server,

fastapi-mcp, LangChain components, and python-dotenv into the .venv` virtual environment.

- MCP tool names are derived from the FastAPI route's operation_id.
- If not specified, FastAPI generates one automatically (e.g., read_user_users__user_id__get).
- It's recommended to set explicit operation_ids on FastAPI routes for clearer MCP tool names:
``python @app.get("/users/{user_id}", operation_id="get_user_info") async def read_user(user_id: int):

- If FastAPI routes are added _after_ mcp.mount()is called, they won't be automatically included. - Solution: Callmcp.setup_server()again after defining the new routes.``python
app = FastAPI()
mcp = FastApiMCP(app)
mcp.mount()

@app.get("/new/endpoint", operation_id="new_tool")
async def new_endpoint(): ...

Claude Desktop / Cursor

Paste into your MCP client config file to install this server.

{
    "mcpServers": {
        "fastapi mcp server + langchain client example": {
            "sse-mcp-and-langchain-client-example": {
                "command": "uv",
                "args": [
                    "init",
                    "#",
                    "If",
                    "pyproject.toml",
                    "doesnt exist"
                ]
            }
        }
    }
}

McpServers

{
    "sse-mcp-and-langchain-client-example": {
        "command": "uv",
        "args": [
            "init",
            "#",
            "If",
            "pyproject.toml",
            "doesnt exist"
        ]
    }
}

> Example project demonstrating how to expose FastAPI endpoints as Model Context Protocol (MCP) tools using fastapi-mcp. Includes a basic LangChain agent (langchain_client.py) that connects to the local FastAPI server via HTTP/SSE using langchain-mcp-adapters to discover and use the exposed tools.

Prerequisites

Getting Started / How to Run

1. Clone the Repository: ``

bash
    git clone <your-repo-url>
    cd <repo-directory>


2.  Install

uv: If you don't have it, install the uv

 package manager (see Project Setup below for command).
3.  Set up Environment & Install Dependencies:

bash
    uv init # If pyproject.toml doesn't exist
    uv venv # Create virtual environment (.venv)
    # Install all project dependencies
    uv pip install fastapi "uvicorn[standard]" fastapi-mcp langchain-mcp-adapters langgraph langchain-openai python-dotenv


4.  Create

.env File: Create a file named .env

 in the project root directory and add your OpenAI API key:

dotenv
    OPENAI_API_KEY=your_openai_api_key_here


5.  Run the FastAPI MCP Server: Open a terminal and run:

bash
    uvicorn main:app --reload --port 8000


    Keep this terminal running.
    _(Alternatively, you can use the

Python: FastAPI MCP debug configuration defined in .vscode/launch.json

 within VS Code / Cursor to run the server with the debugger attached.)_
6.  Run the LangChain Client: Open a _second_ terminal and run:

bash
    uv run python langchain_client.py



    The client will connect to the server, discover tools, and run a query using the agent.

7.  (Optional) Test Server with MCP Inspector: Before running the LangChain client, or for more direct testing, you can use the official MCP Inspector tool:

    - Ensure the FastAPI server is running (Step 5).
    - Open another terminal and run:

npx @modelcontextprotocol/inspector

_(

npx

 comes with Node.js/npm. If this command fails, ensure Node.js is installed and accessible in your PATH.)_
    - In the inspector UI, connect to your server URL:

http://127.0.0.1:8000/mcp


    - Navigate to "Tools", click "List Tools" to see

read_root__get and greet_user_greet__name__get

.
    - Select a tool, fill parameters (e.g.,

name

 for greet_user), and click "Run Tool".

8.  (Optional) Test

greet_user with LangChain Client: The greet_user endpoint and the corresponding test query (query2) in langchain_client.py

 are currently active. Simply run the LangChain client (Step 6) and observe the second part of its execution where it should attempt to greet the user 'LangChain'.
    - _(If you want to disable this test, comment out the

@app.get("/greet/{name}") endpoint in main.py and the query2 section in langchain_client.py

)_

Goal


To build a simple FastAPI server with MCP capabilities for learning and testing purposes, runnable locally and connectable from MCP clients like the Cursor editor's agent.

Project Setup


1.  Package Manager: We used

, a fast Python package installer and resolver written in Rust.
    - Installation (Windows PowerShell):

irm https://astral.sh/uv/install.ps1 | iex


    - Project Initialization:

uv init (Creates pyproject.toml

)
    - Virtual Environment:

uv venv (Creates and manages .venv

)
2.  Dependencies: Installed using

bash
    # Specific commands used during development (covered by the combined install in Getting Started):
    # uv pip install fastapi "uvicorn[standard]" fastapi-mcp
    # uv pip install langchain-mcp-adapters langgraph langchain-openai python-dotenv


    This installs FastAPI, the Uvicorn ASGI server,

fastapi-mcp, LangChain components, and python-dotenv into the .venv

 virtual environment.

`Application (`main.py`)`



A simple FastAPI app was created with endpoints:

-

: Returns a welcome message.
-

/greet/{name}

: Returns a personalized greeting (currently commented out).

Crucially, the

fastapi-mcp

 integration happens after the FastAPI route definitions:

``python from fastapi import FastAPI from fastapi_mcp import FastApiMCP app = FastAPI(...)

No reviews yet — be the first

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.

About

Details

Explore

Setting up with Highlight

From the repository

Prerequisites

Getting Started / How to Run

Project Setup

Claude Desktop / Cursor

McpServers

Prerequisites

Getting Started / How to Run

Goal

Project Setup

Application (main.py)

More like this

Context 7

1Panel

🔎 GPT Researcher

Serena

`Application (`main.py`)`