Simulation MCP

AI tools can operate the Dreapex TMM simulator over MCP. In your AI tool, describe what you want in plain language, and the AI builds the structure, sets the optical parameters, runs the simulation, and reads the results — without opening the web app.

MCP address:

https://tmm.dreapex.com/api/mcp/jsonrpc

What the AI can do

Once connected, you can ask the AI in plain language to:

• Build multilayer structures (single layers, periodic stacks, emissive layers) and set the surrounding media;
• Pick materials from the refractive-index database, or set a constant / custom index;
• Set wavelength, angle, polarization, detectors, incident spectrum, and color;
• Run calculations, parameter sweeps, and optimizations;
• Run Emission simulations;
• Read results, and import / export .tmm models or share links.

Add the Simulation MCP in AI tools

Claude Code

Add the server using the CLI:

claude mcp add --transport http dreapex-tmm https://tmm.dreapex.com/api/mcp/jsonrpc

Cursor

Create or update .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "dreapex-tmm": {
      "type": "http",
      "url": "https://tmm.dreapex.com/api/mcp/jsonrpc"
    }
  }
}

Visual Studio Code

Ensure the GitHub Copilot and GitHub Copilot Chat extensions are installed, then create or update .vscode/mcp.json:

{
  "servers": {
    "dreapex-tmm": {
      "type": "http",
      "url": "https://tmm.dreapex.com/api/mcp/jsonrpc"
    }
  }
}

Windsurf

Go to Settings > Windsurf Settings > Cascade, click Manage MCPs, choose View raw config, and add:

{
  "mcpServers": {
    "dreapex-tmm": {
      "type": "http",
      "url": "https://tmm.dreapex.com/api/mcp/jsonrpc"
    }
  }
}

This file is typically .codeium/windsurf/mcp_config.json.

Zed

Go to Settings > Open Settings and add to the JSON settings file:

{
  "context_servers": {
    "dreapex-tmm": {
      "source": "custom",
      "command": "npx",
      "args": ["mcp-remote", "https://tmm.dreapex.com/api/mcp/jsonrpc"],
      "env": {}
    }
  }
}

This file is typically .config/zed/settings.json.

Authentication

Confirming it works

After setup, just ask the AI to do something, for example:

• Add a 100 nm SiO2 layer on glass and compute reflectance from 400 to 800 nm.
• Build a 7-pair SiO2/TiO2 Bragg mirror and tell me the peak reflectance.

The AI operates the simulator on its own and returns the result.

Relationship to other features

Troubleshooting

Next

• Build and simulate interactively inside the web app: AI Assistant
• Ground the AI in this documentation as well: Docs MCP