Click on "Install 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., "@PyMOL-MCP-VisFetch protein 6M0J, show it as cartoon, and take a screenshot."
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.
PyMOL-MCP-Vis: PyMOL + Claude AI with Visual Feedback
Forked from - Credit to the original author for the core MCP integration.
PyMOL-MCP-Vis extends the original PyMOL-MCP with screenshot capture capabilities, enabling Claude to actually see and analyze molecular structures visually. This allows for true visual feedback loops in structural biology workflows.
What's New in This Fork
Screenshot capture -
take_screenshot(width, height)captures the current PyMOL viewAll-views composite -
take_all_views()captures 6 orthogonal views (front, back, left, right, top, bottom) in a single composite imageVisual feedback loop - Claude can now execute commands AND verify the results visually
Optimized defaults - 200x200px default size balances quality and token efficiency
https://github.com/user-attachments/assets/687f43dc-d45e-477e-ac2b-7438e175cb36
Features
Two-way communication: Connect Claude AI to PyMOL through a socket-based server
Intelligent command parsing: Natural language processing for PyMOL commands
Molecular visualization control: Manipulate representations, colors, and views
Structural analysis: Perform measurements, alignments, and other analyses
Code execution: Run arbitrary Python code in PyMOL from Claude
Installation Guide
Prerequisites
PyMOL installed on your system
Claude for Desktop
Python 3.10 or newer
Git
Step 1: Install the UV Package Manager
On macOS:
On Windows:
For other platforms, visit the UV installation guide.
Step 2: Clone the Repository
Step 3: Set Up the Environment
Create and activate a Python virtual environment:
On macOS/Linux:
On Windows:
Step 4: Install Dependencies
With the virtual environment activated:
Step 5: Configure Claude Desktop
Open Claude Desktop
Go to Claude > Settings > Developer > Edit Config
This will open the
claude_desktop_config.jsonfileAdd the MCP server configuration:
For example:
Note: Use the actual full paths on your system. On Windows, use forward slashes (/) instead of backslashes.
Step 6: Install the PyMOL Plugin
Open PyMOL
Go to Plugin → Plugin Manager
Click on "Install New Plugin" tab
Select "Choose file..." and navigate to the cloned repository
Select the
pymol-mcp-socket-plugin/__init__.pyfileClick "Open" and follow the prompts to install the plugin
Usage
Starting the Connection
In PyMOL:
Go to Plugin → PyMOL MCP Socket Plugin
Click "Start Listening"
The status should change to "Listening on port 9876"
In Claude Desktop:
You should see a hammer icon in the tools section when chatting
Click it to access the PyMOL tools
Example Commands
Here are some examples of what you can ask Claude to do:
"Load PDB 1UBQ and display it as cartoon"
"Color the protein by secondary structure"
"Highlight the active site residues with sticks representation"
"Align two structures and show their differences"
"Calculate the distance between these two residues"
"Save this view as a high-resolution image"
Troubleshooting
Connection issues: Make sure the PyMOL plugin is listening before attempting to connect from Claude
Command errors: Check the PyMOL output window for any error messages
Plugin not appearing: Restart PyMOL and check that the plugin was correctly installed
Claude not connecting: Verify the paths in your Claude configuration file are correct
Limitations & Notes
The socket connection requires both PyMOL and Claude to be running on the same machine
Some complex operations may need to be broken down into simpler steps
Always save your work before using experimental features
Join our Bio-MCP Community to troubleshoot, provide feedback & improve Bio-MCPS! https://join.slack.com/t/bio-mcpslack/shared_invite/zt-31z4pho39-K5tb6sZ1hUvrFyoPmKihAA
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.