pyNastran MCP Server

A Model Context Protocol (MCP) Server for pyNastran, built with FastMCP. Enables AI agents to interact with Nastran FEA models.

Features

• 🔧 BDF Tools: Read, write, and analyze Nastran input files
• 📊 OP2 Tools: Extract results from Nastran output files
• 🔍 Geometry Tools: Mesh quality checks and geometric analysis
• 📝 Analysis Tools: Automated report generation
• 🚀 FastMCP: Built with modern FastMCP framework
• 🌐 Multiple Transports: stdio, SSE, and streamable-http

Installation

pip install pynastran-mcp

Or install from source:

git clone https://github.com/Shaoqigit/pynastran-mcp.git
cd pynastran-mcp
pip install -e .

Quick Start

Stdio Transport (Default)

For MCP clients like Cherry Studio, Claude Desktop:

pynastran-mcp

SSE Transport

# Default: host=127.0.0.1, port=8080
pynastran-mcp --transport sse

# Custom host and port
pynastran-mcp --transport sse --host 0.0.0.0 --port 8080

Streamable HTTP Transport (Production)

# Default: host=127.0.0.1, port=8080
pynastran-mcp --transport streamable-http

# Custom host and port
pynastran-mcp --transport streamable-http --host 0.0.0.0 --port 8080

MCP Client Configuration

Cherry Studio / Cursor / Claude Desktop

Add to your MCP client configuration:

{
  "mcpServers": {
    "pynastran": {
      "command": "pynastran-mcp"
    }
  }
}

See CHERRY_STUDIO_TUTORIAL.md for detailed setup instructions.

Available Tools

BDF Tools

Tool	Description
read_bdf	Read BDF file and return model summary
get_model_info	Get detailed model information
write_bdf	Write model to new BDF file
get_nodes	Get node coordinates
get_elements	Get element connectivity
get_materials	Get material properties
get_properties	Get property definitions

OP2 Tools

Tool	Description
read_op2	Read OP2 result file
get_result_cases	List available result cases
get_stress	Extract stress results
get_displacement	Extract displacement results

Geometry Tools

Tool	Description
check_mesh_quality	Check mesh quality metrics
get_model_bounds	Get model bounding box

Analysis Tools

Tool	Description
generate_report	Generate comprehensive analysis report

Usage Examples

With AI Agents

Once configured, you can ask your AI assistant:

"Read the BDF file at /path/to/model.bdf and tell me about the mesh"
"Analyze the stress results from /path/to/results.op2"
"Check the mesh quality and suggest improvements"
"Generate a report for my Nastran model"

Programmatic Usage

from pynastran_mcp.tools.bdf_tools import BdfTools
from pynastran_mcp.tools.op2_tools import Op2Tools

async def analyze_model():
    # BDF Analysis
    bdf_tools = BdfTools()
    summary = await bdf_tools.read_bdf("wing.bdf")
    print(summary)
    
    # OP2 Results
    op2_tools = Op2Tools()
    stresses = await op2_tools.get_stress("results.op2", element_type="CQUAD4")
    print(stresses)

Project Structure

pynastran-mcp/
├── pynastran_mcp/
│   ├── __init__.py
│   ├── server.py          # FastMCP server with all tools
│   └── tools/
│       ├── __init__.py
│       ├── bdf_tools.py   # BDF file operations
│       ├── op2_tools.py   # OP2 result operations
│       ├── geometry_tools.py  # Mesh quality checks
│       └── analysis_tools.py  # Report generation
├── pyproject.toml
├── README.md
└── examples/
    └── example_usage.py

Requirements

• Python 3.10+
• pyNastran >= 1.4.0
• mcp >= 1.0.0 (with FastMCP)

Development

# Setup
git clone https://github.com/Shaoqigit/pynastran-mcp.git
cd pynastran-mcp
pip install -e ".[dev]"

# Run tests
pytest

# Code formatting
black pynastran_mcp/

License

MIT License - see LICENSE file

Acknowledgments

• pyNastran - The underlying Nastran interface library
• MCP Python SDK - FastMCP framework

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Please make sure to update tests as appropriate and follow the existing code style.

• pyNastran - The underlying Nastran interface library
• MCP Python SDK - FastMCP framework