About Fermatmcp
Fermat MCP is a FastMCP server for numerical and symbolic mathematics, along with plotting. Built on NumPy, SymPy and Matplotlib, it suits users who want an AI assistant to run mathematical operations and generate charts.
Key features
- 01NumPy-based numerical operations
- 02SymPy-based symbolic algebra and calculus
- 03Matplotlib chart and equation plotting
- 04Linear algebra such as eigenvalues, solve and SVD
- 05Calculus operations including diff, integrate and limit
Use cases
- 01Computing eigenvalues of a matrix from an AI assistant
- 02Plotting functions such as x**2 and sin(x) together
- 03Simplifying expressions or solving equations symbolically
MCP support
This project is an MCP server that exposes mathematical computation and plotting operations as tools to MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEFermat MCP
This project provides a FastMCP server for mathematical computations, including numerical and symbolic calculations, as well as plotting.
Modules
1. mpl_mcp - Matplotlib Integration
| Feature | Description |
|---|---|
plot_barchart | Plots bar charts of given data values |
plot_scatter | Creates scatter plots from data points |
plot_chart | Plots line, scatter, or bar charts |
plot_stem | Creates stem plots for discrete data |
plot_stack | Generates stacked area/bar charts |
eqn_chart | Plots mathematical equations |
2. numpy_mcp - NumPy Integration
| Category | Operations |
|---|---|
| Basic Math | add, sub, mul, div, power, abs, exp, log, sqrt |
| Trigonometric | sin, cos, tan |
| Statistics | mean, median, std, var, min, max, argmin, argmax, percentile |
| Linear Algebra | dot, matmul, inv, det, eig, solve, svd |
| Matrix Operations | create, zeros, ones, full, arange, linspace |
| Array Manipulation | reshape, flatten, concatenate, transpose, stack |
3. sympy_mcp - SymPy Integration
| Category | Operations |
|---|---|
| Algebra | simplify, expand, factor, collect |
| Calculus | diff, integrate, limit, series |
| Equations | solve, solveset, linsolve, nonlinsolve |
| Matrix Operations | create, det, inv, rref, eigenvals |
Setup
Requirements
-
Python 3.12 or higher (To install Python3.12 follow Python Download)
-
uv (To install uv follow uv Installation)
Clone the repository
git clone https://github.com/abhiphile/fermat-mcp
Visual Studio Code, Windsurf
You can find the mcp.json file in the
MCP: Open User Configuration or MCP: Open Workspace Configuration
Add the following to your mcp.json:
{
"mcpServers": {
"fmcp": {
"command": "bash",
"args": ["MCP_SERVER_ABSOLUTE_PATH/setup.sh"],
"description": "fmcp server is for mathematical computations, including numerical and symbolic calculations, as well as plotting."
}
}
}
Claude (Anthropic)
If you're using Claude or the Anthropic MCP client, add this working MCP configuration to your mcp.json (update the directory path to your local clone):
{
"mcpServers": {
"fmcp": {
"command": "uv",
"args": [
"--directory",
"/home/ty/Repositories/fermat-mcp",
"run",
"server.py"
]
}
}
}
Gemini CLI
-
Open your Gemini settings JSON located in ~/.gemini/settings.json where ~ is your home directory.
-
Add the following to your settings.json:
{
"mcpServers": {
"fmcp": {
"command": "bash",
"args": ["MCP_SERVER_ABSOLUTE_PATH/setup.sh"],
"description": "fmcp server is for mathematical computations, including numerical and symbolic calculations, as well as plotting."
}
}
}
Installing via Smithery
To install Fermat MCP for local usage automatically via Smithery:
npx -y @smithery/cli install @abhiphile/fermat-mcp --client gemini
Example Usage
- Using Gemini CLI
╭──────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ > Can you use fmcp server and using numpy method find the eigen values of this 8*8 matrix, |
│ 2 1 3 1 1 8 4 2 |
│ 6 6 0 7 1 4 6 1 │
│ 9 2 1 8 7 9 9 0 │
│ 2 5 6 6 9 8 0 1 │
│ 1 3 6 2 3 8 8 1 │
│ 9 4 2 2 1 2 2 9 │
│ 8 6 4 4 2 0 2 8 │
│ 0 0 0 6 6 7 5 6 │
╰──────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ ✔ numpy_mcp_numerical_operation (fmcp MCP Server) {"a":[[2,1,3,1,1,8,4,2],[6,6,0,7,1,4,6,1],[9,2,1,8,7,9,9,0],[2,5,6,6,9,8,0,1],[1,3,… │
│ │
│ {"eigenvalues":["32.077244457548815+0j","-11.531090644775198+0j","-6.6653982146786195+0j","0.6715984762411508+3.37024850 │
│ 10270413j","0.6715984762411508-3.3702485010270413j","4.541270555490195+2.776364664923869j","4.541270555490195-2.77636466 │
│ 4923869j","3.6935063384423428+0j"],"eigenvectors":[["-0.23263835483680192+0j","-0.2264723575289234+0j","-0.4308391916391 │
│ 0195+0j","-0.012346573390129022+0.17748655663058255j","-0.012346573390129022-0.17748655663058255j","-0.21221572277027187 │
│ +0.3524396218277479j","-0.21221572277027187-0.3524396218277479j","0.3451499664861578+0j"],["-0.31955742545335186+0j","-0 │
│ .2569860493445581+0j","0.05691886770041556+0j","-0.35591013681869693-0.2242364092694275j","-0.35591013681869693+0.224236 │
│ 4092694275j","0.1932161673963751-0.39527849111641133j","0.1932161673963751+0.39527849111641133j","-0.7979681696063214+0j │
│ "],["-0.46626263247473404+0j","-0.4684914620112376+0j","0.5469400556350749+0j","0.34325164099973565+0.06607019711949293j │
│ ","0.34325164099973565-0.06607019711949293j","0.21312270185159682+0.28822307710358636j","0.21312270185159682-0.288223077 │
│ 10358636j","0.42707422750984786+0j"],["-0.41589316441674523+0j","0.2291771012892302+0j","0.09410792992600435+0j","0.6375 │
│ 92441360358+0j","0.637592441360358+-0j","0.46446646137729414+0j","0.46446646137729414+-0j","0.08171661775583623+0j"],["- │
│ 0.35812884189789035+0j","-0.26551071423139044+0j","-0.649979374400915+0j","-0.2999153430497845+0.20110182336747695j","-0 │
│ .2999153430497845-0.20110182336747695j","0.04424367733961273-0.020400357434930304j","0.04424367733961273+0.0204003574349 │
│ 30304j","-0.037276545325188024+0j"],["-0.3042416233100104+0j","0.4027405071320744+0j","0.2343447346668298+0j","-0.119224 │
│ 87219398595+0.10317726668781271j","-0.11922487219398595-0.10317726668781271j","-0.3387759141362585+0.15790241128043445j" │
│ ,"-0.3387759141362585-0.15790241128043445j","0.11473144194899779+0j"],["-0.35389732755448017+0j","0.5359362131480812+0j" │
│ ,"0.13989902586074168+0j","-0.16008513226461316-0.26111873108374495j","-0.16008513226461316+0.26111873108374495j","0.055 │
│ 729749650881726-0.22748184206919145j","0.055729749650881726+0.22748184206919145j","-0.19838514696566653+0j"],["-0.327615 │
│ 5213691367+0j","-0.30122843153433887+0j","0.07858579556736099+0j","0.00042251379361638556-0.11670182367986544j","0.00042 │
│ 251379361638556+0.11670182367986544j","-0.18590228871569267-0.2479130526139182j","-0.18590228871569267+0.247913052613918 │
│ 2j","-0.03374593856285834+0j"]]} │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
✦ Of course, I can do that. Here are the eigenvalues for the given matrix:
32.077244457548815
-11.531090644775198
-6.6653982146786195
0.6715984762411508 + 3.3702485010270413j
0.6715984762411508 - 3.3702485010270413j
4.541270555490195 + 2.776364664923869j
4.541270555490195 - 2.776364664923869j
3.6935063384423428
- Using Visual Studio Code, Windsurf
Here the task was to plot both x**2 and sin(x) in the same plot.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.