Summary
NumPy, SymPy and Matplotlib reachable as tools instead of as a scratch script.
The symbolic half is what matters: an integral or an eigenvalue comes back computed by SymPy rather than recalled, and the plotting tools then draw whatever you just worked out.
What it is
A FastMCP server in three parts: Matplotlib plotting, NumPy numeric work, and SymPy symbolic algebra and calculus.
What you get
- Plots: line, scatter and bar charts, stem plots for discrete data, stacked area and bar charts
- Mathematical equations plotted directly
- NumPy operations: arithmetic, trigonometry, statistics, linear algebra, matrix creation, array reshaping
- SymPy algebra: simplify, expand, factor, collect
- Calculus: differentiate, integrate, limits, series
- Equation solving: solve, solveset, linsolve, nonlinsolve
- Matrix work in both libraries — determinant, inverse, eigenvalues, rref, svd
Requirements
Python 3.12 or higher and uv. Cloned from the repository and launched through its setup.sh from the client config.
Setup effort
One command — npx -y @smithery/cli install @abhiphile/fermat-mcp --client gemini