MCP Server Exploration

🚀 MCP Introduction

A simple example of a client and server using the Connect API.

🚀 Quick Start

Run an MCP Server

CONNECT_API_KEY="Your API Key" SWAGGER_FILE="swagger.yaml" make server

Run an MCP Client

make client

You can modify the chat section in the main function of app.py to ask different questions.

💻 Usage Examples

Basic Usage

The following is an example code snippet showing how to register Swagger tools with Chatlas:

import chatlas
import requests
from openapi_mcp.chatlas import SwaggerTool
from openapi_mcp.connect_api import map_operations_to_tools
from openapi_mcp.swagger import (
    expand_all_references,
    transform_swagger_to_operation_dict,
)

api_url = "127.0.0.1:8000/"
openapi_dict = requests.get(f"{api_url}/swagger.json").json()

openapi_doc = expand_all_references(openapi_dict)
operations = transform_swagger_to_operation_dict(openapi_doc)
tools = map_operations_to_tools(operations)

aws_model = os.getenv("AWS_MODEL", "us.anthropic.claude-3-5-sonnet-20241022-v2:0")
aws_region = os.getenv("AWS_REGION", "us-east-1")
chat = chatlas.ChatBedrockAnthropic(model=aws_model, aws_region=aws_region)
for operation in api_operations.values():
    SwaggerTool.register_tool(
        chat,
        SwaggerTool(
            base_url=api_url,
            operation=operation,
        ),
    )

Advanced Usage

To see the actual effect, run a locally compatible OpenAPI server and start the Shiny application:

# Run a locally compatible OpenAPI server
make ex-starwars

# Start Shiny
make shiny

📚 Documentation

Task Description

• [x] Read the Swagger file.
• [x] Expand the reference objects in the Swagger file: doc = expand_swagger(doc).
• [x] Create a bridge between the Swagger file and tool calls.
• [x] Register the tools with Chatlas.
• [x] Create a function to process generic Swagger files: operations = transform_swagger_to_operation_dict(doc).
• [x] Move all mcp_servers to the openapi_mcp method.
• [x] Create a Shiny chat application.
  • [x] Accept the plumber API URL.
  • [x] Display the Swagger documentation on the home page.
  • [x] Utilize the new functions.
• [x] Package it as a library?
  • [x] Export a function that accepts Swagger information and returns tool calls (as shown above).