- Home
- MCP servers
- Chuk MCP Math Server
Chuk MCP Math Server
- python
4
GitHub Stars
python
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
You have access to a high-performance Mathematical Computation Protocol (MCP) server that exposes a large library of mathematical functions through flexible transport options. It delivers fast tool calls, robust error handling, and zero-config startup, making it ideal for integrating advanced math capabilities into local workflows or web-based applications.
How to use
You can use the MCP server by connecting your client to either a remote HTTP endpoint or a local STDIO transport. For web-style usage, you connect via HTTP to the public or a self-hosted server that implements the MCP interface and request mathematical operations by name and arguments. For local development or desktop automation, you run a local MCP server instance and communicate through STDIO, which provides instant access to the complete function library.
How to install
Prerequisites you need before installation: Python 3.11 or newer, and a modern environment capable of running Python packages. You may also use a transport helper to simplify running the server locally.
# Option 1: Run a remote public server (no installation required for the client)
# Use a client that supports MCP and point to the public endpoint:
# https://math.chukai.io/mcp
Additional setup and commands
If you choose to run locally, you typically install the package and start the local server with a convenient command. The following patterns show common ways to run the server via STDIO or HTTP transports.
# Install locally (example using Python package manager)
pip install chuk-mcp-math-server
# Run as a local STDIO server (default)
uv run chuk-mcp-math-server
# Run as an HTTP server on port 8000
uv run chuk-mcp-math-server --transport http --port 8000
Configuration overview
You have multiple ways to configure the MCP server, including command line options, configuration files, and environment variables. You can filter which functions are available, set computation timeouts, and tune performance. Defaults are sensible for zero-configuration startup, but you can customize behavior as needed.
Public server usage
A free public math server is available for quick testing and learning. You can connect Claude Desktop or any MCP client to the public endpoint to access all 642 functions.
Available tools
add
Add two numbers. Uses standard arithmetic addition and supports integer and floating point inputs.
is_prime
Check whether a given integer is prime.
fibonacci
Compute the nth Fibonacci number or generate a Fibonacci sequence.
gcd
Calculate the greatest common divisor of two integers.
matrix_multiply
Multiply two matrices, returning the resulting matrix.
dot_product
Compute the dot product of two vectors.
distance
Compute the distance between two points in space.
circle_area
Calculate the area of a circle given its radius.
triangle_area
Compute the area of a triangle given base and height.
normal_pdf
Evaluate the probability density of a normal distribution.
binomial_pmf
Calculate the probability mass function of a binomial distribution.
mean
Compute the arithmetic mean of a dataset.
median
Compute the median of a dataset.
std_dev
Compute the standard deviation of a dataset.
variance
Compute the variance of a dataset.
derivative
Compute the derivative of a function at a point or symbolically.
integrate
Compute the definite or indefinite integral of a function.
newton_raphson
Find roots of a function using the Newton-Raphson method.
gcd_extended
Extended GCD algorithm providing coefficients for Bézout's identity.
factorial
Compute the factorial of a non-negative integer.