该项目演示了如何将亚马逊Bedrock上的基础模型与Chainlit和MCP(模式上下文协议)服务器相融合,以创建具有工具增强能力的交互式聊天界面.

□ 项目概览

此样本应用程序显示:

• 使用Anthropic Claude 3.7 Sonnet与亚马逊·贝德罗克融合
• 用于对话的 AI 互动的 Chainlit 网络界面
• 执行基本数学操作的自定义模式背景协议服务器
• 为顺利的用户体验而流出答复
• LangChain与MCP适配器的集成,连接外部工具
• LangGraph 用于创建一个指挥工具使用的ReAct(Reasoning and Action)代理人

□ 先决条件

在开始之前,请务必:

• ** AWS账户**: 你需要一个AWS的账户 与贝德洛克访问
• Bedrock Access:确保您的AWS账户能够访问Anthropic Claude 3.7 Sonnet模型
• [AWS CLI**: 配置有适当的证书和权限
• ** 第3.13分部分: 这个项目需要 Python 3. 13 或更新
• UV:依赖管理和建筑(安装在[T0])

□ 安装

1. 清除储存库:

2. 联合国.

Amazon Bedrock and MCP Server integration, the easy way

This project demonstrates how to integrate foundation models on Amazon Bedrock with Chainlit and MCP (Model Context Protocol) servers to create an interactive chat interface with tool-enhanced capabilities.

Project Overview

This sample application showcases:

• Integration with Amazon Bedrock using Anthropic Claude 3.7 Sonnet
• A Chainlit web interface for conversational AI interactions
• A custom Model Context Protocol (MCP) server that performs basic math operations
• Streaming responses for a smooth user experience
• LangChain integration with MCP adapters to connect external tools
• LangGraph for creating a ReAct (Reasoning and Acting) agent that orchestrates tool use

Prerequisites

Before getting started, ensure you have:

• AWS Account: You’ll need an AWS account with Bedrock access
• Bedrock Access: Ensure your AWS account has access to the Anthropic Claude 3.7 Sonnet model
• AWS CLI: Configured with appropriate credentials and permissions
• Python 3.13+: This project requires Python 3.13 or newer
• UV: For dependency management and building (install with pip install uv)

Installation

1. Clone the repository:

   git clone https://github.com/aws-samples/sample-bedrock-mcp.git
   cd sample-bedrock-chainlit-mcp
   

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

   uv sync --all-groups
   

3. Configure AWS credentials (if not already done):

   aws configure
   

Building from Source

To build the project from source using UV:

1. Build the package:

   uv build
   

This will create distribution files in the dist/ directory, including the wheel file dist/sample_bedrock_chainlit_mcp-0.1.0-py3-none-any.whl that can be installed or used directly as the MCP server.

Usage Guide

Starting the Chainlit Application

To start the Chainlit application:

chainlit run src/ui/app.py

This will launch a web server (typically at ) with the chat interface.

Setting up the MCP Server

The Math MCP server needs to be connected to provide calculation capabilities:

1. In the Chainlit web interface, look for the small plug icon under the chat text input element
2. Click on it to open the MCP connection dialog
3. Enter the following details:
   • Name: math
   • Command: uvx dist/sample_bedrock_chainlit_mcp-0.1.0-py3-none-any.whl
   • Type: default (stdio)
4. Click “Confirm”

A successful connection will be indicated in the interface, and the math tools will become available to the Claude model.

Example Usage

Once the Chainlit application is running and the MCP server is connected, you can ask mathematical questions like:

• “What’s (3 + 5) x 12?”
• “Can you calculate 144 divided by 12?”
• “If I have 7 apples and get 9 more, then give 4 away, how many do I have left?”