MCPServerDemo

This project implements a simple JSON-RPC server using FastAPI, Uvicorn, and jsonrpcserver in Python. It exposes add and multiply methods.

Setup

1. Clone the repository:

   git clone <your-repo-url>
   cd json-rpc-server
   

2. Create a virtual environment and install dependencies using uv:

   uv venv
   .\.venv\Scripts\activate # On Windows
   # source .venv/bin/activate # On Linux/macOS
   uv pip install fastapi uvicorn jsonrpcserver
   

Running the Server

To start the development server with auto-reloading:

.\.venv\Scripts\activate # On Windows
# source .venv/bin/activate # On Linux/macOS
uv run python -m uvicorn main:app --reload

The server will run at http://127.0.0.1:8000/.

API Endpoints

GET /

A simple test endpoint to confirm the server is running.

Request:

# Using curl (Linux/macOS/Git Bash)
curl http://127.0.0.1:8000/

# Using Invoke-RestMethod (PowerShell)
Invoke-RestMethod -Uri "http://127.0.0.1:8000/" -Method Get

Response:

{
  "message": "FastAPI is running!"
}

POST / (JSON-RPC 2.0)

This is the main JSON-RPC endpoint for add and multiply methods.

Add Method

Request:

# Using curl
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc": "2.0", "method": "add", "params": [2, 3], "id": 1}' http://127.0.0.1:8000/

# Using Invoke-RestMethod (PowerShell)
Invoke-RestMethod -Uri "http://127.0.0.1:8000/"   -Method Post -Headers @{ "Content-Type" = "application/json" }   -Body '{"jsonrpc":"2.0","method":"add","params":[2,3],"id":1}'

Response (Example):

{"jsonrpc": "2.0", "result": 5, "id": 1}

Multiply Method

Request:

# Using curl
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc": "2.0", "method": "multiply", "params": [4, 5], "id": 1}' http://127.0.0.1:8000/

# Using Invoke-RestMethod (PowerShell)
Invoke-RestMethod -Uri "http://127.0.0.1:8000/"  -Method Post -Headers @{ "Content-Type" = "application/json" }  -Body '{"jsonrpc":"2.0","method":"multiply","params":[4,5],"id":1}'

Response (Example):

{"jsonrpc": "2.0", "result": 20, "id": 1}

Multipl Methods in a sigal call

curl -X POST -H "Content-Type: application/json" -d '[{"jsonrpc": "2.0", "method": "add", "params": [2, 3], "id": 1},{"jsonrpc": "2.0", "method": "multiply", "params": [2, 30], "id": 2}]' http://127.0.0.1:8000/


Invoke-RestMethod -Uri "http://127.0.0.1:8000/" -Method Post -Headers @{ "Content-Type" = "application/json" }  -Body '[{"jsonrpc":"2.0","method":"add","params":[2,3],"id":1},{"jsonrpc":"2.0","method":"multiply","params":[2,30],"id":2}]'

Install fastmcp

uv add fastmcp

Run MCP inspector

uv run fastmcp dev inspector TestMPCserver.py 

Run MCP server

uv run fastmcp run TestMPCserver.py

Connect to cluade-desktop

uv run fastmcp install claude-desktop TestMPCserver.py

Run MCP server

uv run fastmcp run DataprovidersMCPServer.py

Connect to cluade-desktop

uv run fastmcp install claude-desktop DataprovidersMCPServer.py

uv add langchain langchain-openai langchain-mcp-adapters python-dotenv streamlit logging pandas fastmcp python-dotenv jsonrpcserver uvicorn streamlit langchain-ollama fastapi

 uvicorn serversendevent:app

 uvicorn Streamable:app

Streaming HTTP Demo

This project includes a streaming HTTP endpoint demonstration using FastAPI's StreamingResponse.

Start the Stream Server

uv run uvicorn Streamable:app --reload

The server will run at http://127.0.0.1:8000/stream and sends 10 chunks with 1-second intervals.

Run the Stream Client

python StreamableClient.py

The client demonstrates both async and synchronous methods to consume the streaming response.

Test with curl

# Using curl (Linux/macOS/Git Bash)
curl http://127.0.0.1:8000/stream

# Using Invoke-RestMethod (PowerShell)
Invoke-RestMethod -Uri "http://127.0.0.1:8000/stream" -Method Get

VS Code Debugging

A launch.json file is provided in the .vscode directory to enable debugging with VS Code. You can set breakpoints in main.py and run the "Python: FastAPI" configuration.