
用Claude和SnapTrade搭建一个交易机器人。
概览
该GitHub仓库展示了一个演示项目,将MCP(模型上下文协议)服务器集成到SnapTrade,实现与Anthropic的Claude AI连接。它展示了如何通过MCP框架向Claude开放通过SnapTrade访问的金融账户数据。 这个概念验证展示了一个基本的流程: - SnapTrade 集成:连接 SnapTrade 的 API 以访问金融账户信息。- MCP 服务器实现:通过模型上下文协议暴露 SnapTrade 功能和数据。- Claude 连接性:允许 Claude 或其他兼容 MCP 的 AI 模型与 MCP 服务器暴露的数据交互。 目标是提供一个简单的示例,展示如何构建MCP服务器,将AI模型与第三方API(如SnapTrade)连接起来。 - 注册SnapTrade:您必须创建自己的SnapTrade账户以获得独特的客户ID和客户秘密。
概述(AI 翻译)
该GitHub仓库展示了一个演示项目,将MCP(模型上下文协议)服务器集成到SnapTrade,实现与Anthropic的Claude AI连接。它展示了如何通过MCP框架向Claude开放通过SnapTrade访问的金融账户数据。
项目概述:
这个概念验证展示了一个基本的流程:
- SnapTrade 集成:连接 SnapTrade 的 API 以访问金融账户信息。
- MCP 服务器实现:通过模型上下文协议(Model Context Protocol)暴露 SnapTrade 功能和数据。
- Claude 连接性:允许 Claude 或其他兼容 MCP 的 AI 模型与 MCP 服务器暴露的数据交互。
目标是提供一个简单的示例,展示如何构建MCP服务器,将AI模型与第三方API(如SnapTrade)连接起来。
重要设置说明
要运行本次演示,您需要:
- 注册SnapTrade:您必须创建自己的SnapTrade账户以获得独特的客户ID和客户秘密。
- 注册演示用户:利用SnapTrade的API文档演示页面上的用户注册流程。这将为你提供必要的用户凭证。
本页由 AI 翻译,技术细节请以原文为准。
README
Demo: MCP Server for SnapTrade Integration with Claude AI
This GitHub repository showcases a demonstration project for integrating an MCP (Model Context Protocol) server with SnapTrade, enabling a connection to Anthropic’s Claude AI. It illustrates how financial account data, accessed via SnapTrade, can be made available to Claude through the MCP framework.
Project Overview:
This proof-of-concept demonstrates a basic pipeline:
- SnapTrade Integration: Connects to SnapTrade’s API to access financial account information.
- MCP Server Implementation: Exposes SnapTrade functionalities and data through the Model Context Protocol.
- Claude Connectivity: Allows Claude, or other MCP-compatible AI models, to interact with the data exposed by the MCP server.
The goal is to provide a simple example of how an MCP server can be built to bridge AI models with third-party APIs like SnapTrade.
Important Setup Instructions
To run this demo, you will need to:
- Sign up for SnapTrade: You must create your own SnapTrade account to obtain a unique Client ID and Client Secret.
- Register a Demo User: Utilize the user registration process available on SnapTrade’s API documentation demo page. This will provide you with the necessary user credentials.
- Configure Environment Variables: Once you have your SnapTrade Client ID, Client Secret, and registered user details, populate these values in the
.envfile located in the build output folder of this project. See the.env.examplefile for an example.
You will also need to configure your Claude Desktop app to load the new MCP server, like so:
{
"mcpServers": {
"SnapTrade": {
"command": "/home/username/projects/mcp-snaptrade/bin/cli"
}
}
}
Please Note:
- USE AT YOUR OWN RISK
- Demo Purposes Only: This project is intended as a demonstration and a learning tool. It is not designed for production use and may not cover all edge cases or possess robust error handling.
- Personal Project & Likely Unmaintained: This repository is a personal project. As such, it is unlikely to receive regular updates, bug fixes, or active maintenance. Please use it as-is with this understanding.
I hope this example provides a helpful starting point for understanding how MCP servers can facilitate connections between AI models and external data services like SnapTrade!
安装
This server does not publish a one-line install command.
Open the repository installation guide 新手?先看 mcpskill.org 教程 →