Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@lys-mcp load iris.csv and make a scatter plot of sepal length vs width"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
lys_mcp enables Claude Code and other processes to control lys, an interactive multi-dimensional data analysis and visualization platform. With lys_mcp, Claude can execute Python commands in lys and see the graphs drawn by lys and matplotlib.
Claude Code ⇄ (MCP) ⇄ lys-mcp ⇄ (local socket) ⇄ lys (python -m lys --remote)Check out the documentation for more information.
git clone https://github.com/lys-devel/lys_mcp.git
cd lys_mcp
pip install -e ".[mcp]"
claude mcp add --scope user lys -- lys-mcpThen launch lys in remote mode:
python -m lys --remoteTo control lys on another computer in the same network, launch lys with --port (e.g. python -m lys --remote --port 8765) and connect by tcp://<host>:8765. See usage.
See the install documentation and usage for details.
Related MCP server: ipython-kernel
Please read CONTRIBUTING.md.
This project is licensed under the GPLv3 License - see the LICENSE.md file for details
This server cannot be deployed
Use your Mac, Windows or Linux computer from ChatGPT, Claude or Codex: files, commands, documents.
Work on your own computers from Claude or ChatGPT: run commands, edit and search files.
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Real-time chat for AI agents. Claude Code, Cursor, Cline and Codex join channels over MCP.
