MCP App with Authentication

MCP App with Authentication

This MCP server implements a secure OAuth 2.1 authorization server with Google login, enabling authenticated tool execution and a user interface for MCP applications.

Category
访问服务器

README

MCP App with Authentication

This project implements a secure, database-backed OAuth 2.1 authorization server and MCP App on top of FastMCP. It allows clients like Claude, ChatGPT, or the MCP Inspector to sign in using Google OAuth, execute authorized tools, and display details in an MCP App UI.


Setup & Running Guide

1. Ngrok Setup & Static URL Configuration

  1. Install Ngrok: Follow the setup and installation instructions on the official ngrok Setup & Installation page.

  2. Configure your Static Domain: To keep your public OAuth redirect URL persistent and avoid updating the Google Cloud Console settings every time you restart ngrok, configure a static development domain in your local ngrok.yml file:

    Run the ngrok configuration editor:

    ngrok config edit
    
  3. Add Endpoint Configuration: Paste the following configuration structure into the editor, replacing authtoken with your actual token from the ngrok dashboard:

    version: 3
    agent:
      authtoken: YOUR_NGROK_AUTHTOKEN
    
    endpoints:
      - name: local
        url: <your-static-domain>.ngrok-free.app
        upstream:
          url: 8000
    
  4. Save and close the editor.

  5. Launch the static tunnel:

    ngrok start local
    

Keep this tunnel active. Your public URL will always be https://<your-static-domain>.ngrok-free.app.


2. Google Cloud Console Setup

  1. Create a Google Cloud Project: Go to the Google Cloud Console and create a new project.
  2. Configure OAuth Consent Screen:
    • Go to APIs & Services > OAuth consent screen.
    • Choose External user type and fill out the required app information.
    • Add scopes: .../auth/userinfo.email, .../auth/userinfo.profile, and openid.
    • Add your Google account as a test user.
  3. Create Credentials:
    • Go to APIs & Services > Credentials.
    • Click Create Credentials > OAuth client ID.
    • Choose Web application as the application type.
    • Under Authorized JavaScript origins, add:
      • https://<your-static-domain>.ngrok-free.app
    • Under Authorized redirect URIs, add:
      • https://<your-static-domain>.ngrok-free.app/auth/google/callback (Our server callback)
  4. Copy Secrets:
    • Save the client ID and client secret. These will be assigned to the environment variables GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET at server launch.

3. Start the MCP Server

Create a .env file in the root of the project to store your configurations and Google credentials securely:

PUBLIC_URL="https://<your-static-domain>.ngrok-free.app"
GOOGLE_CLIENT_ID="your-google-client-id.apps.googleusercontent.com"
GOOGLE_CLIENT_SECRET="your-google-client-secret"

Once your .env file is set up, start the server:

uv run python main.py

4. Build the Vite UI (React + Tailwind CSS)

To compile the React + Tailwind single-file HTML bundle to ui/dist/index.html:

cd ui
bun run build

5. Start the MCP Inspector

Run the local MCP Inspector using Bun (bunx) or NPM (npx):

# Using Bun (bunx)
bunx @modelcontextprotocol/inspector --transport http --server-url https://<your-static-domain>.ngrok-free.app/mcp

# Or using NPM (npx)
npx @modelcontextprotocol/inspector --transport http --server-url https://<your-static-domain>.ngrok-free.app/mcp

Open the generated local URL, click Connect, and complete the Google login flow to launch the MCP App UI.


References

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选