MCP Everything

🚀 MCP All-in-One Server

This MCP server attempts to showcase all the features of the MCP protocol. It is not intended to be a practical server but rather a demonstration server created for the purpose of testing MCP client builders. It implements functions such as prompts, tools, resources, sampling, etc., demonstrating the capabilities of MCP.

⚠️ Important Note

This project is extracted from https://github.com/modelcontextprotocol/servers/tree/main/src/everything to create an independent implementation.

🚀 Quick Start

📦 Installation

Local Installation

# Clone the repository
git clone https://github.com/modelcontextprotocol/mcp-everything.git
cd mcp-everything

# Install dependencies
npm install

# Build the project
npm run build

# Start the server
npm start

Global Installation

# Install globally from npm
npm install -g mcp-everything

# Run the server
mcp-everything

Docker

# Build the Docker image
docker build -t mcp-everything .

# Run the container
docker run -it mcp-everything

💻 Usage Examples

Using with Claude Desktop

Add the following content to claude_desktop_config.json:

{
  "mcpServers": {
    "everything": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-everything"
      ]
    }
  }
}

📚 Documentation

✨ Features

Tools

1. echo

   • A simple tool that echoes the input message.
   • Input:
     • message (string): The message to be echoed.
   • Return: The text content containing the echoed message.

2. add

   • Adds two numbers.
   • Input:
     • a (number): The first number.
     • b (number): The second number.
   • Return: A text string of the addition result.

3. longRunningOperation

   • Demonstrates the notification function for long-running operations.
   • Input:
     • duration (number, default: 10): The duration in seconds.
     • steps (number, default: 5): The number of progress steps.
   • Return: A completion message containing the duration and the number of steps.
   • Sends progress notifications during execution.

4. sampleLLM

   • Demonstrates the LLM sampling ability using the MCP sampling function.
   • Input:
     • prompt (string): The prompt sent to the LLM.
     • maxTokens (number, default: 100): The maximum number of generated tokens.
   • Return: The response generated by the LLM.

5. getTinyImage

   • Returns a small test image.
   • No input required.
   • Return: Base64-encoded PNG image data.

6. printEnv

   • Prints all environment variables.
   • Used for debugging the MCP server configuration.
   • No input required.
   • Return: A JSON string of the environment variables.

7. annotatedMessage

   • Demonstrates how to use annotations to provide metadata for content.
   • Input:
     • messageType (enum: "error" | "success" | "debug"): The type of message to be demonstrated.
     • includeImage (boolean, default: false): Whether to include an example image.
   • Return: The message content with annotations.

Resources

• Contains 100 test resources, divided into two categories:
  • Even-numbered: Text-type resources.
  • Odd-numbered: Binary blob-type resources.
• Supports the following functions:
  • Paged browsing.
  • Resource subscription and update notifications.

Prompts

• Contains two prompt templates:
  • Simple Prompt: A basic single-round conversation prompt.
  • Complex Prompt: A multi-round conversation prompt that requires parameters (such as temperature) and an optional style.

Logs

• Sends a random-level log message every 15 seconds to test the client's log reception ability.