Elysia MCP Plugin

Elysia MCP Plugin

A comprehensive ElysiaJS plugin for building Model Context Protocol (MCP) servers with HTTP transport, session management, and support for tools, resources, and prompts.

Category
Visit Server

README

Elysia MCP Plugin

A comprehensive ElysiaJS plugin for implementing Model Context Protocol (MCP) servers with HTTP transport support.

Features

  • HTTP Transport: Full HTTP-based MCP transport with Streamable HTTP
  • Session Management: Stateful session handling via headers
  • Type-Safe: Built with TypeScript and Zod validation
  • Easy Integration: Simple plugin architecture for Elysia apps
  • Comprehensive Support: Tools, Resources, Prompts, and Logging
  • Error Handling: Proper JSON-RPC 2.0 error responses
  • Testing: Full unit test coverage with Bun test runner

Installation

bun add elysia-mcp
# or
npm install elysia-mcp

Starter Template

To quickly get started with a pre-configured Elysia MCP project, you can use our starter template:

# Create a new project from the starter template
bun create https://github.com/kerlos/elysia-mcp-starter my-mcp-project

# Navigate to the project
cd my-mcp-project

# Install dependencies
bun install

# Start development server
bun run dev

The elysia-mcp-starter template includes:

  • Pre-configured Elysia setup with MCP plugin
  • TypeScript configuration
  • Development scripts
  • Basic project structure
  • Example MCP server implementation

Quick Start

import { Elysia } from 'elysia';
import { mcp, McpServer } from 'elysia-mcp';
import { z } from 'zod';

const app = new Elysia()
  .use(
    mcp({
      serverInfo: {
        name: 'my-mcp-server',
        version: '1.0.0',
      },
      capabilities: {
        tools: {},
        resources: {},
        prompts: {},
        logging: {},
      },
      setupServer: async (server: McpServer) => {
        // Register your MCP tools, resources, and prompts here
        server.tool(
          'echo',
          {
            text: z.string().describe('Text to echo back'),
          },
          async (args) => {
            return {
              content: [{ type: 'text', text: `Echo: ${args.text}` }],
            };
          }
        );
      },
    })
  )
  .listen(3000);

Usage

Running the Examples

Basic Example:

# Run the basic example server (port 3000)
bun run example

# Or with development mode (auto-restart)
bun run dev

Multiple Servers Example:

# Run the multiple MCP servers example (port 3000)
bun example:multi

This example demonstrates how to create multiple MCP plugins in a single Elysia application:

  • Math Operations Plugin (/math) - Basic arithmetic tools:

    • add - Add two numbers
    • multiply - Multiply two numbers
    • power - Calculate base to the power of exponent
  • Text Utilities Plugin (/text) - Text processing tools:

    • uppercase - Convert text to uppercase
    • word_count - Count words in text
    • reverse - Reverse text characters
    • replace - Replace text with global matching

Testing with MCP Inspector

  1. Install MCP Inspector:

    npx @modelcontextprotocol/inspector
    
  2. Connect to your server:

    • Basic Example: http://localhost:3000/mcp
    • Multiple Servers Example:
      • Math Plugin: http://localhost:3000/math
      • Text Plugin: http://localhost:3000/text

Configuration Options

  • serverInfo: Server information
  • capabilities: MCP capabilities to advertise
  • enableLogging: Enable debug logging (default: false)
  • setupServer: Callback to register tools, resources, and prompts

Session Management

The plugin automatically handles session management via the Mcp-Session-Id header. Each session maintains its own state and can be terminated cleanly.

Modular Handler Architecture

The plugin supports a modular handler architecture that allows you to create specialized endpoints for different MCP capabilities:

import {
  mcp,
  ToolsHandler,
  ResourcesHandler,
  PromptsHandler,
} from 'elysia-mcp';

const app = new Elysia().use(
  mcp({
    /* config */
  })
);

API Reference

Tools

Register tools using the MCP Server instance:

server.tool(
  'tool-name',
  {
    param: z.string().describe('Parameter description'),
  },
  async (args) => {
    // Tool implementation
    return {
      content: [{ type: 'text', text: 'Tool result' }],
    };
  }
);

Resources

Register resources for file or data access:

server.resource('Resource Name', 'resource://uri', async () => {
  return {
    contents: [
      {
        uri: 'resource://uri',
        mimeType: 'text/plain',
        text: 'Resource content',
      },
    ],
  };
});

Prompts

Register reusable prompt templates following MCP best practices:

server.prompt(
  'prompt-name',
  'Prompt description',
  {
    param: z.string().describe('Parameter description'),
  },
  async (args) => {
    return {
      description: 'Generated prompt',
      messages: [
        {
          role: 'user',
          content: {
            type: 'text',
            text: `Generated prompt with ${args.param}`,
          },
        },
      ],
    };
  }
);

Testing

Run the comprehensive test suite:

bun test

License

MIT - see LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Related

Plugin Configuration

Plugin Options

interface MCPPluginOptions {
  serverInfo?: {
    name: string;
    version: string;
  };
  capabilities?: ServerCapabilities;
  enableLogging?: boolean;
  setupServer?: (server: McpServer) => void | Promise<void>;
}

Architecture

┌─────────────────┐    ┌──────────────┐    ┌─────────────────┐
│   HTTP Client   │───▶│ Elysia HTTP  │───▶│    MCP Plugin   │
│                 │    │   Handler    │    │                 │
└─────────────────┘    └──────────────┘    └─────────────────┘
                                                     │
                                                     │
                                            ┌─────────────────┐
                                            │    McpServer    │
                                            │   (Singleton)   │
                                            └─────────────────┘

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured