
MCP Think
A Model Context Protocol server that provides AI assistants like Claude with a dedicated space for structured thinking during complex problem-solving tasks.
README
MCP Think
A Model Context Protocol (MCP) server implementing the "think" tool for improving Claude's and other LLMs' complex reasoning capabilities.
Overview
This MCP server implements the "think" tool as described in Anthropic's blog post, which provides Claude with a dedicated space for structured thinking during complex problem-solving tasks. The think tool has been shown to significantly improve performance in complex tasks requiring policy adherence and reasoning in long chains of tool calls.
Features
- Structured Thinking Space: Provides LLMs with a dedicated place to break down complex problems
- Thought History: Maintains a log of all thoughts with timestamps for reference
- Multiple Transport Support: Works with both stdio and SSE transports
Installation
Install from PyPI:
pip install mcp-think
Or install from source:
git clone https://github.com/ddkang1/mcp-think.git
cd mcp-think
pip install -e .
Usage
You can run the MCP server directly:
mcp-think
By default, it uses stdio transport. To use SSE transport:
mcp-think --sse
You can also specify host and port for SSE transport:
mcp-think --sse --host 127.0.0.1 --port 3001
Configuration
To use this tool with Claude in Windsurf, add the following configuration to your MCP config file:
"think": {
"command": "/home/xxx/.local/bin/mcp-think",
"args": [],
"type": "stdio",
"pollingInterval": 30000,
"startupTimeout": 30000,
"restartOnFailure": true
}
The command
field should point to the directory where you installed the python package using pip.
Available Tools
The MCP server provides the following tool:
- think: Record a thought with a timestamp
Development
Installation for Development
git clone https://github.com/ddkang1/mcp-think.git
cd mcp-think
pip install -e ".[dev]"
Running Tests
pytest
Code Style
This project uses Black for formatting, isort for import sorting, and flake8 for linting:
black src tests
isort src tests
flake8 src tests
Contributing
Contributions are welcome! Please see CONTRIBUTING.md for details.
Changelog
See CHANGELOG.md for a history of changes to this project.
License
This project is licensed under the MIT License - see the LICENSE file for details.
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。