Think Tool Integration for Claude AI

A minimal MCP Server based on Anthropic's "think" tool research.

Overview

This project implements a minimal Message Control Protocol (MCP) server that provides Claude AI models with the "think" tool capability. Based on Anthropic's research published on March 20, 2025, this implementation enables Claude to perform better on complex reasoning tasks involving multi-step tool usage.

What is the "think" Tool?

The "think" tool gives Claude the ability to include an additional thinking step—with its own designated space—as part of reaching a final answer. Unlike extended thinking (which happens before response generation), the "think" tool allows Claude to pause during response generation to consider whether it has all necessary information to proceed.

Key Benefits

• Improves complex problem-solving performance
• Enhances policy adherence in tool usage
• Increases consistency in decision making
• Helps with multi-step problems requiring careful reasoning

Implementation

This server implements the "think" tool with the following specification:

{
  "name": "think",
  "description": "Use the tool to think about something. It will not obtain new information or change the database, but just append the thought to the log. Use it when complex reasoning or some cache memory is needed.",
  "input_schema": {
    "type": "object",
    "properties": {
      "thought": {
        "type": "string",
        "description": "A thought to think about."
      }
    },
    "required": ["thought"]
  }
}

When to Use the "think" Tool

Based on Anthropic's research, this tool is most beneficial for:

1. Tool Output Analysis: When Claude needs to process previous tool call outputs before acting
2. Policy-Heavy Environments: When Claude must follow detailed guidelines
3. Sequential Decision Making: When each action builds on previous ones and mistakes are costly

Implementation Best Practices

Strategic Prompting

For best results, include clear instructions in your prompts on when and how to use the "think" tool. Consider providing domain-specific examples that show:
- Expected detail level in reasoning
- How to break down complex instructions into steps
- Decision trees for common scenarios
- Information verification processes

System Prompt Integration

Complex guidance works best when placed in the system prompt rather than the tool description itself.

How It Works

The server operates using the Model Context Protocol (MCP) to communicate with Claude and similar AI assistants. It:
- Runs as a standalone process using stdio for communication
- Registers the "think" tool for Claude to use during reasoning
- Returns structured responses that can be processed by AI assistants
- Logs thinking steps without affecting the external environment

Features

Tools

• think: Enables Claude to think about a problem or analyze information
• Required: thought (string containing Claude's thinking process)

Development

Install dependencies:

npm install

Build the server:

npm run build

For development with auto-rebuild:

npm run watch

Debugging

Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector:

npm run inspector

The Inspector will provide a URL to access debugging tools in your browser.

Installation

Installing via npm

npm install -g think-mcp-server

Claude Desktop

Add the server config at:
- MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%/Claude/claude_desktop_config.json

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

Cline

1. Open the Cline extension settings
2. Open "MCP Servers" tab
3. Click on "Configure MCP Servers"
4. Add the server config:

{
  "mcpServers": {
    "github.com/marcopesani/think-mcp-server": {
      "command": "npx",
      "args": ["-y", "think-mcp-server"],
      "disabled": false,
      "autoApprove": ["think"]
    }
  }
}

Cursor

1. Open the Cursor settings
2. Open "Features" settings
3. In the "MCP Servers" section, click on "Add new MCP Server"
4. Choose a name, and select "command" as "Type"
5. In the "Command" field, enter the following:

npx -y think-mcp-server

Docker

Build the image:

docker build -t think-mcp-server .

Run the container:

docker run -it think-mcp-server

For development, mount your source code as a volume:

docker run -v $(pwd):/app think-mcp-server

Getting Started

1. Install the server using one of the methods above
2. Configure your AI client to use the think-mcp-server
3. In your prompts to Claude, include instructions for using the "think" tool
4. For best results, add examples of effective thinking patterns in your system prompt

TypeScript Development Example

Here's an example prompt focused on TypeScript development to help Claude leverage the "think" tool effectively: