home / mcp / axiomatic mcp server
Provides access to Axiomatic tools for scientific computing, document processing, and photonics design via an MCP client.
Configuration
View docs{
"mcpServers": {
"axiomatic-ai-ax-mcp": {
"command": "uvx",
"args": [
"--from",
"axiomatic-mcp",
"all"
],
"env": {
"AXIOMATIC_API_KEY": "your-api-key-here"
}
}
}
}You can run Axiomatic MCP Server locally to give AI assistants access to the Axiomatic platform’s tools for scientific computing, document processing, and photonics design. This MCP server configuration lets you start a standardized, secure bridge between your client and the Axiomatic API so you can use all available tools from your preferred MCP client.
Launch and connect your MCP client to the Axiomatic MCP Server. Use the single, standard configuration to run the MCP server locally, then add or remove tools as needed. Once the server is running, your MCP client will expose the full set of Axiomatic capabilities, such as equation exploration, document parsing, annotation, photonics previews, data extraction from plots, and model fitting.
Prerequisites you need before installation:
Step-by-step commands to install and run the MCP server locally:
# 1) Install Python (if not already installed)
# 2) Install uvx, the MCP runtime
# You can typically install uvx via your Python environment:
# pip install uvx
# 3) Obtain your API key and store it securely. You will need this when configuring the MCP server.
# 4) Start the MCP server using the standard configuration below (no need to modify other settings):Configuration, security, and troubleshooting details are provided here to help you run the MCP server smoothly and securely.
Environment variables you will use and what they control are shown in the server configuration snippet. The key variable shown here is AXIOMATIC_API_KEY, which must be set to your API key to access Axiomatic services.
Compose equations of interest by combining information from scientific papers and domain knowledge, enabling fast exploration of mathematical models.
Convert PDF documents to markdown with advanced OCR and layout understanding to preserve structure and context.
Create intelligent annotations for PDFs with contextual analysis, equation extraction, and parameter identification.
Design photonic integrated circuits from natural language descriptions and simulate concepts before prototyping.
Extract numerical data from plot images for quantitative analysis and reproducibility.
Fit parametric models or digital twins to observational data using advanced statistical techniques and optimization.