# clouatre-labs/math-mcp-learning-server [Health: Active]

**Category:** 🧮 Data Science Tools  
**Repository:** https://github.com/clouatre-labs/math-mcp-learning-server  
**GitHub Stars:** 5  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/clouatre-labs-math-mcp-learning-server

## Description
Educational MCP server for math operations, statistics, visualization, and persistent workspaces. Built with FastMCP 2.0.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "math-mcp-learning-server": {
    "url": "https://math-mcp.fastmcp.app/mcp"
  }
}
```

## Documentation

## What clouatre-labs/math-mcp-learning-server MCP server does

The clouatre-labs/math-mcp-learning-server MCP server adds a collection of educational mathematics capabilities to an MCP client. Its 17 tools cover expression evaluation, statistical summaries, compound-interest calculations, unit conversion, matrix operations, plotting, and workspace persistence. The server is built with FastMCP and the official MCP Python SDK.

Workspace tools can save calculations and load them later. The server also publishes MCP resources for a workspace summary, calculation history, available functions, mathematical constants, tool metadata, active variables, and a health check. Two prompts are included: `math_tutor` for structured tutoring at a configurable difficulty, and `formula_explainer` for step-by-step formula explanations.

## How it works

An MCP client sends tool calls, reads resources, or invokes prompts exposed by the server. Mathematical expression evaluation is restricted to an allowlist of characters and functions, with access limited to the `math` module and `abs`; dangerous built-ins and imports are not available. Tool inputs use Pydantic validation, and file operations are limited to the designated workspace directory.

The calculation tools include statistical measures such as mean, median, mode, standard deviation, and variance. Matrix tools handle multiplication, transposition, determinants, inverses, and eigenvalues. Visualization tools generate function plots, histograms, line charts, scatter plots, box plots, and financial line charts.

## Setup and configuration

The clouatre-labs/math-mcp-learning-server MCP server can be used as a hosted service without local installation. Its HTTP MCP endpoint is `https://math-mcp.fastmcp.app/mcp`. The README provides a Claude Desktop configuration using the HTTP transport and this URL.

For local execution, configure an MCP client to run `uvx` with the PyPI package and the scientific and plotting extras:

```json
{
  "mcpServers": {
    "math": {
      "command": "uvx",
      "args": ["math-mcp-learning-server[scientific,plotting]"]
    }
  }
}
```

The project is also published on PyPI and listed in the MCP Registry. Other package configurations are documented by the project for basic, scientific-only, and plotting-only installations.

## Tools and capabilities

- `workspace_save` and `workspace_load` persist and retrieve calculations.
- `calc_expression` evaluates supported mathematical expressions.
- `calc_statistics`, `calc_interest`, and `calc_units` handle statistics, compound interest, and conversions between length, weight, and temperature units.
- Matrix tools perform common linear algebra operations, including eigenvalue calculation.
- Plotting tools create seven types of mathematical, statistical, and financial charts.
- Resources expose workspace state, history, functions, constants, variables, tool metadata, and health information.
- `math_tutor` and `formula_explainer` provide structured educational prompts.

## Limitations and notes

The server is designed for educational math operations rather than general-purpose code execution. Expression evaluation is intentionally constrained, and file access is restricted to the designated workspace directory. The README does not specify client compatibility beyond the provided Claude Desktop configuration, so support for other clients should be verified separately. Cloud and local usage paths are both documented, but the material does not describe authentication requirements for the hosted endpoint.

The project uses the Apache-2.0 license. Its repository documentation includes separate guides for architecture, examples, cloud deployment, development, and contribution workflows.

_Full upstream README: https://allmcps.com/mcp/clouatre-labs-math-mcp-learning-server/readme_