MCP Scratchpad Server

An MCP (Model Context Protocol) server that provides a "think" tool for structured reasoning, based on the Anthropic blog post.

Overview

The Scratchpad MCP server allows Claude and other LLMs to add dedicated thinking steps during complex tool use scenarios. This improves problem-solving by creating space for structured reasoning and helps the model carefully process tool call outputs.

Related MCP server: Sequential Thinking MCP Server

Features

• Think Tool: Add thoughts to a timestamped log for structured reasoning

• Thoughts Log Resource: View all logged thoughts in JSON format

• Clear Tool: Reset the thinking log when needed

• Structured Thinking Prompt: Template for systematic problem-solving

Installation

Quick Start (Recommended)

Run directly with uvx (no installation needed):

uvx mcp-server-scratchpad

PyPI Installation

Install globally with pip or uv:

# Using pip
pip install mcp-server-scratchpad

# Using uv
uv pip install mcp-server-scratchpad

Development Installation

For local development:

# Clone the repository
git clone https://github.com/JoshuaOliphant/scratchpad_mcp.git
cd scratchpad_mcp

# Install dependencies
uv sync

# Run the server
uv run mcp-server-scratchpad

Usage

With Claude Desktop

Add to your Claude Desktop configuration:

On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

Using uvx:

{
  "mcpServers": {
    "scratchpad": {
      "command": "uvx",
      "args": ["mcp-server-scratchpad"]
    }
  }
}

Using pip installation:

{
  "mcpServers": {
    "scratchpad": {
      "command": "mcp-server-scratchpad"
    }
  }
}

Testing with MCP Inspector

For development and testing:

# Using uvx
uvx mcp-server-scratchpad

# Using MCP Inspector (if developing locally)
uv run mcp dev src/mcp_server_scratchpad/server.py

Tools

think

Add a thinking step to the log.

Parameters:

• thought (string): The thought to log

Example:

{
  "thought": "I need to break this problem into smaller parts..."
}

clear_thoughts

Clear all thoughts from the log.

No parameters required

Resources

log://thoughts

Returns all logged thoughts as a JSON array with timestamps.

Example output:

[
  {
    "timestamp": "2024-01-20T10:30:45.123456",
    "thought": "First, I need to understand the requirements..."
  },
  {
    "timestamp": "2024-01-20T10:31:02.789012",
    "thought": "The main components are X, Y, and Z..."
  }
]

Prompts

Structured Thinking

A template for systematic problem-solving using the think tool.

Parameters:

• topic (string): The topic or problem to think through

Use Cases

The scratchpad tool is particularly useful for:

1. Long chains of tool calls: Keep track of intermediate results and reasoning

2. Complex problem-solving: Break down problems systematically

3. Policy-heavy environments: Verify compliance with guidelines

4. Sequential decision making: Document the reasoning process

5. Edge case handling: Think through unusual scenarios

Example Workflow

1. Start with a complex problem

2. Use the "Structured Thinking" prompt to set up the approach

3. Use the think tool multiple times to log reasoning steps

4. Check progress with the log://thoughts resource

5. Clear the log when starting a new problem

Best Practices

• Use descriptive thoughts that explain the "why" not just the "what"

• Break complex problems into smaller, manageable parts

• Validate reasoning at each step

• Summarize conclusions at the end

• Clear the log between unrelated tasks

Development

To modify the server:

1. Fork the repository

2. Make your changes to src/mcp_server_scratchpad/server.py

3. Test with uv run mcp dev src/mcp_server_scratchpad/server.py

4. Submit a pull request

Future Enhancements

Potential improvements:

• Persist thoughts to a file

• Add search/filter capabilities

• Support for thought categories or tags

• Export thoughts in different formats

• Integration with other reasoning tools

License

MIT License - see LICENSE file for details

Credits

Based on the thinking tool concept from Anthropic's blog post