plumed2

The PLUMED2 MCP Server provides AI-assisted capabilities to generate, validate, and optimize PLUMED input files for molecular dynamics simulations. It exposes a set of tools that let you create complete PLUMED inputs from high-level requirements, verify syntax and references, optimize performance for large systems, and retrieve predefined templates for common simulation scenarios. You can use these tools to streamline PLUMED workflows inside AI assistants or other MCP-enabled applications. The server is designed to be invoked via the MCP client configuration and supports a modular set of actions that can be combined in prompts or automated pipelines.

Key tools include generate_plumed_input for creating full PLUMED input content with specified collective variables, bias methods, and output settings; validate_plumed_syntax to check syntax and catch typical errors; optimize_plumed_performance to provide practical tuning suggestions for large-scale simulations; and get_plumed_template to fetch ready-to-use templates tailored to system types or study goals. Templates cover common scenarios such as protein folding, drug binding, phase transitions, and nucleation, helping accelerate template-driven workflows. You can also leverage the templates as starting points and then adjust parameters as needed within the MCP environment.

Prerequisites: - Python 3.11 or higher - pip (bundled with Python) - Access to the repository with the plumed_mcp_server code (or cloned clone of the MCP server repository)

Installation steps:

1. Set up a Python virtual environment (recommended):

python -m venv venv
source venv/bin/activate  # Mac/Linux
venv\Scripts\activate     # Windows

2. Navigate to the server directory (example path shown in README):

cd plumed_mcp_server

3. Install dependencies:

pip install -r requirements.txt

4. Run or configure the MCP server as needed (see your deployment instructions). If testing locally, ensure the server module can be invoked with:

python -m plumed_mcp_server.main

5. Optional: Run tests (if provided by the project):

cd plumed_mcp_server
python -m pytest tests/ -v

Environment and configuration tips:

• Ensure the path in the mcp_config (cwd) points to your actual plumed_mcp_server installation directory.
• If you use a virtual environment, activate it before running the MCP server so dependencies resolve correctly.
• The alwaysAllow list defines which tools are accessible through the MCP client without explicit permission prompts; adjust as needed for your security model.
• If PLUMED or Python paths differ in your environment, consider setting explicit PATH or PYTHONPATH variables.
• For large-scale simulations, provide system_info in requests to enable better optimization guidance from the server.