pyvista
PyVista-based MCP (Model Context Protocol) server
claude mcp add --transport stdio pyvista-pyvista-mcp-server <path-to-python-executable> <path-to-server-script>
How to use
This PyVista-based MCP server exposes a toolset that leverages PyVista for 3D visualization within the MCP framework. The repository describes a hello_world tool that generates a 3x3 grid of 3D shapes (such as cylinders, arrows, spheres, etc.) and exports the result as an interactive HTML file named a_basic.html. You can invoke this tool through the MCP client, which communicates with the server over stdio as configured in the mcpServers section. After running, locate the a_basic.html file in the current working directory and open it in a web browser to explore the interactive visualization. The server is designed to demonstrate how PyVista visuals can be produced and exported in an MCP workflow, integrating 3D rendering with the standard Model Context Protocol tooling.
How to install
Prerequisites:
- Python 3.8+ installed on your system
- Basic familiarity with MCP and tool invocation
Installation steps:
- Clone or download the repository containing the PyVista MCP server
- Create a virtual environment (optional but recommended): python -m venv venv source venv/bin/activate # on Windows use: venv\Scripts\activate
- Install required Python packages (adjust to your environment):
pip install pyvista numpy
If there are additional dependencies listed in requirements.txt, install them as well
- Ensure the server script is executable and note the path to the Python interpreter and the server script:
Example aliases (adjust paths accordingly)
python_path=/path/to/python server_script=/path/to/server_script.py - Prepare the MCP configuration (see mcp_config) and ensure the server can be started via: ${python_path} ${server_script}
- Run the MCP client or tool launcher that uses the above configuration to start the server and communicate over stdio.
Additional notes
Notes and tips:
- The hello_world tool exports an HTML file named a_basic.html. Ensure the working directory is writable and that you have permission to create files there.
- If you encounter issues starting the server, verify the Python environment has PyVista and its dependencies installed (PyVista may require VTK).
- The configuration example uses a placeholder path to the Python executable and the server script; replace these with absolute paths on your system.
- Transport type is stdio, which means the MCP client and server communicate through standard input/output streams; ensure there are no conflicting processes that might intercept these streams.
- If you extend the server with more tools, follow the same pattern: expose tool commands via the MCP protocol and document input/output expectations for each tool.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP