MCP Server for Bun and Elysia
LocalAbout MCP Server for Bun and Elysia
An MCP server implementation built with Bun and the Elysia web framework, using an SSE transport to expose resources, tools and prompts to LLMs. It is intended for developers who want to build MCP servers in TypeScript on the Bun runtime.
Key features
- 01Server-Sent Events transport for Bun and Elysia
- 02Resources, tools and prompts support
- 03Runs on the Bun JavaScript runtime
- 04TypeScript type definitions included
- 05Hot reload in development and production build
Use cases
- 01Building an MCP server on Bun and Elysia
- 02Debugging an SSE-based MCP server with MCP Inspector
MCP support
This project is an MCP server that exposes resources, tools and prompts to MCP clients over an SSE transport.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEMCP Server for Bun and Elysia
An implementation of the Model Context Protocol (MCP) server using Bun and the Elysia web framework. This project enables you to create high-performance MCP servers that expose resources, tools, and prompts to LLMs through a standardized interface.
Features
- Server-Sent Events (SSE) transport implementation for Bun and Elysia
- Complete MCP protocol support with resources, tools, and prompts
- High-performance thanks to Bun's JavaScript runtime
- TypeScript support with proper type definitions
- Easy-to-use API for creating MCP-compatible servers
Prerequisites
- Bun installed on your system
- Basic familiarity with TypeScript and Elysia
Installation
# Clone the repository
git clone <your-repo-url>
cd mcp-server
# Install dependencies
bun install
Usage
Starting the server
# Start the server
bun start
# Start with hot reloading for development
bun dev
Building for production
# Build for production
bun run build
This will create a minified Node.js-compatible build in the dist directory.
Development
Project Structure
src/index.ts- Main entry point for the serversrc/SSEElysiaTransport.ts- SSE transport implementation for Bun and Elysia
Creating an MCP Server
import { McpServer, ResourceTemplate } from "@modelcontextprotocol/sdk/server/mcp.js";
import { z } from "zod";
import { SSEElysiaTransport } from "./SSEElysiaTransport";
import { Elysia } from "elysia";
// Create MCP server
const server = new McpServer({
name: "my-mcp-server",
version: "1.0.0"
});
// Add resources, tools, and prompts
server.resource(
"example",
"example://resource",
async (uri) => ({
contents: [{
uri: uri.href,
text: "Example resource content"
}]
})
);
// Create Elysia app
const app = new Elysia()
.get("/", () => "MCP Server")
.get("/sse", async (context) => {
try {
// Create transport
const transport = new SSEElysiaTransport("/messages", context);
// Store transport
const sessionId = transport.sessionId;
// ... store transport in a map
// Connect to MCP server
await server.connect(transport);
return;
} catch (error) {
// Handle error
}
})
.post("/messages", async (context) => {
// Handle incoming messages
});
// Start server
app.listen(3001, () => {
console.log("MCP Server running at http://localhost:3001");
});
Debugging
You can debug your MCP server using the MCP Inspector tool and connect through sse
npx @modelcontextprotocol/inspector
This will open a web interface where you can:
- List available resources, tools, and prompts
- Test calling tools and retrieving resources
- Inspect the communication between the client and server
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.