jupyter-mcp-server

概述

由Datalayer开发的Jupyter MCP服务器允许AI模型实时连接和管理Jupyter笔记本。它提供了一套全面的工具用于笔记本管理、单元格操作和执行，支持多模态输出和JupyterLab集成。该服务器能够在整个笔记本范围内保持上下文感知，并与Claude Desktop、Cursor和Windsurf等任何MCP客户端协作。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• list_files

  List files and directories in the Jupyter server's file system.

  试试:Show me the files in the current directory试试:List all notebooks in the project
• list_kernels

  List all available and running kernel sessions on the Jupyter server.

  试试:Show all running Jupyter kernels试试:List available Python kernels
• connect_to_jupyter

  Connect to a Jupyter server dynamically without restarting the MCP server.

  试试:Connect to the Jupyter server at localhost:8888试试:Establish connection to production Jupyter
• use_notebook

  Connect to a notebook file, create a new one, or switch between notebooks.

  试试:Open my_analysis.ipynb notebook试试:Create a new notebook called results
• list_notebooks

  List all notebooks available on the Jupyter server and their status.

  试试:Show all available notebooks试试:List notebooks and their running status
• restart_notebook

  Restart the kernel for a specific managed notebook.

  试试:Restart the kernel for my_analysis.ipynb试试:Reset the notebook environment
• unuse_notebook

  Disconnect from a specific notebook and release its resources.

  试试:Close the current notebook试试:Disconnect from my_analysis.ipynb
• read_notebook

  Read notebook cells source content with brief or detailed format options.

  试试:Show me all cells in my_analysis.ipynb试试:Display notebook content with outputs
• read_cell

  Read the full content (Metadata, Source and Outputs) of a single cell.

  试试:Show me cell 5 contents试试:Display the content of the selected cell
• insert_cell

  Insert a new code or markdown cell at a specified position.

  试试:Add a new code cell after cell 3试试:Insert a markdown cell at the beginning
• delete_cell

  Delete a cell at a specified index.

  试试:Remove cell 7试试:Delete the selected cell
• execute_cell

  Execute a cell with timeout, supports multimodal output including images.

  试试:Run cell 4试试:Execute the selected cell and show results

可对比工具

安装

# 安装必需的依赖
pip install jupyterlab==4.4.1 jupyter-collaboration==4.0.2 jupyter-mcp-tools>=0.1.4 ipykernel pycrdt

# 启动JupyterLab
jupyter lab --port 8888 --IdentityProvider.token MY_TOKEN --ip 0.0.0.0

{
  "mcpServers": {
    "jupyter": {
      "command": "uvx",
      "args": ["jupyter-mcp-server@latest"],
      "env": {
        "JUPYTER_URL": "http://localhost:8888",
        "JUPYTER_TOKEN": "MY_TOKEN",
        "ALLOW_IMG_OUTPUT": "true"
      }
    }
  }
}

FAQ

哪些Jupyter部署与此MCP服务器兼容？

该服务器兼容任何Jupyter部署，包括本地服务器、JupyterHub和云平台。它也与Datalayer托管的笔记本兼容。

服务器如何处理笔记本上下文感知？

服务器通过理解完整的笔记本结构、单元格关系和执行历史来维护整个笔记本范围内的上下文感知，从而提供更相关的AI交互。

jupyter-mcp-server 对比