EdgeOne Pages: MCP Client and Server Implementation with Functions

What is EdgeOne Pages: MCP Client and Server Implementation with Functions?

This project is an intelligent chat application built with EdgeOne Pages Functions technology. It implements a complete Model Context Protocol (MCP) workflow, enabling browser-based access to MCP tool capabilities such as generating online webpages from a single prompt.

How to use EdgeOne Pages: MCP Client and Server Implementation with Functions?

Configure a remote MCP service by adding "edgeone-pages-mcp-server" with the URL https://mcp-on-edge.edgeone.site/mcp-server to your MCP client settings. For local development, install dependencies with npm install, set environment variables in .env, and start the dev server with npm run dev. The application can also be deployed directly via the EdgeOne Pages deploy button.

Key features of EdgeOne Pages: MCP Client and Server Implementation with Functions

• Interactive chat interface built with Next.js and React
• High‑performance edge functions on EdgeOne Pages
• Complete MCP implementation based on latest Streamable HTTP spec
• Backend API compatible with OpenAI request/response format
• Browser‑based access to powerful MCP tool capabilities

Use cases of EdgeOne Pages: MCP Client and Server Implementation with Functions

• Generating online webpages with a single prompt using MCP tools
• Building intelligent chat applications with edge‑deployed serverless functions
• Testing and integrating MCP Streamable HTTP servers
• Prototyping OpenAI‑compatible backends with context‑aware workflows

FAQ from EdgeOne Pages: MCP Client and Server Implementation with Functions

What is the architecture of the project?

The system consists of an MCP Streamable HTTP Server, an MCP Client, and a Backend API that coordinates the MCP workflow as the MCP HOST.

How do I configure the AI service?

Set your AI service interface credentials in a .env file by copying .env.example and filling in the values before starting the development server.

Does the backend support OpenAI‑format requests?

Yes, the backend API at functions/v1/chat/completions fully handles OpenAI‑formatted request and response payloads.

What transport protocol does the MCP server use?

The MCP server is based on the Streamable HTTP transport, as defined in the 2025‑03‑26 version of the Model Context Protocol specification.

How can I deploy my own instance?

Click the “Deploy with EdgeOne Pages” badge in the README, or manually configure an EdgeOne Pages project with the provided template.