# abhiphile/fermat-mcp [Health: Active]

**Category:** 🧮 Data Science Tools  
**Repository:** https://github.com/abhiphile/fermat-mcp  
**GitHub Stars:** 20  
**npm Downloads (last month):** 33418  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/abhiphile-fermat-mcp

## Description
The ultimate math engine unifying SymPy, NumPy & Matplotlib in one powerful server. Perfect for developers & researchers needing symbolic algebra, numerical computing, and data visualization.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "fermat-mcp": {
    "command": "npx",
    "args": ["-y","@smithery/cli"]
  }
}
```

## Documentation

## What abhiphile/fermat-mcp MCP server does

The abhiphile/fermat-mcp MCP server provides a local FastMCP interface for mathematical work. Its functionality is divided into three areas: NumPy-backed numerical operations, SymPy-backed symbolic operations, and Matplotlib-backed plotting. The repository describes it as supporting both numerical and symbolic calculations, with chart generation available through the plotting module.

The server can help an MCP client evaluate common mathematical expressions, work with arrays and matrices, solve equations, manipulate symbolic expressions, and produce visual output. It is intended for developers, researchers, and other users who want an agent to call established Python mathematics libraries through MCP rather than perform every operation in conversation.

## How it works

abhiphile/fermat-mcp runs as a local process and is configured as an MCP server in the client’s JSON settings. The documented setup uses Python 3.12 or later and uv. A client starts the repository’s `server.py` through uv, or invokes the supplied `setup.sh` script from a local clone.

The NumPy module covers elementary arithmetic, powers, absolute values, exponentials, logarithms, square roots, trigonometric functions, descriptive statistics, linear algebra, array creation, and array transformations. Matrix-oriented operations include products, inverses, determinants, eigenvalues, solving systems, and singular value decomposition.

The SymPy module handles symbolic simplification, expansion, factoring, collection, differentiation, integration, limits, series, equation solving, and selected matrix operations. Matplotlib functions create bar, scatter, line, stem, stacked, and equation plots.

## Setup and configuration

Clone the repository and install Python 3.12 or newer together with uv. For Visual Studio Code or Windsurf, add an `mcpServers` entry whose command is `bash` and whose argument points to the local clone’s `setup.sh`. The README also documents a Claude or Anthropic configuration that runs `uv --directory <local-directory> run server.py`; the directory must be replaced with the path where the repository was cloned.

Gemini CLI uses a similar `bash` configuration pointing to `setup.sh`, placed in `~/.gemini/settings.json`. The README also lists Smithery as an installation route, but that is a client-specific installer rather than the server’s own runtime command.

## Tools and capabilities

The documented modules provide these capability groups:

- Numerical arithmetic, trigonometry, statistics, array construction, reshaping, concatenation, stacking, and transposition.
- Linear algebra including determinants, inverses, eigenvalues, equation solving, and SVD.
- Symbolic algebra, calculus, equation solving, and matrix manipulation.
- Bar, scatter, line, stem, stacked, and equation charts.

## Limitations and notes

The supplied material does not identify a published package name, hosted endpoint, authentication method, or software license. The examples assume local filesystem access and require users to substitute their own repository path in client configuration. Plotting is described by function and chart type, but the material does not specify an output transport or file format. The README also does not document support for clients beyond the configurations it names.

The abhiphile/fermat-mcp MCP server should therefore be evaluated as a locally run Python tool rather than as a remote mathematical API. Its documented scope is limited to the listed NumPy, SymPy, and Matplotlib operations.

_Full upstream README: https://allmcps.com/mcp/abhiphile-fermat-mcp/readme_