Vibe Math

Vibe Math MCP is a high-performance MCP server that exposes a rich suite of mathematical tools to accelerate computations when interacting with language models. Built to speed up basic arithmetic through advanced calculus and linear algebra, it helps you perform precise calculations, data transformations, and statistical analyses as part of your workflow with LLMs.

How to use

You connect to the Vibe Math MCP server using an MCP client or IDE integration. The server exposes a set of mathematical tools across six domains, allowing you to perform expressions evaluation, statistics, matrix operations, derivatives, integrals, batch orchestration, and more. Use natural language prompts or targeted tool invocations to perform calculations, then feed results back into your model prompts for richer insights.

How to install

Prerequisites: ensure you have a compatible environment for running MCP servers. In most setups you will need a command-line environment and access to the runtime that your chosen MCP client expects. Install steps below describe the local development and published-package usage paths.

1. Prepare your environment. Install the runtime you will use to launch MCP servers (examples show common CLI tools used for MCP workflows). Ensure you can run commands from your terminal.

2. Add the Math MCP server to your MCP client configuration for published-package usage.

{
  "mcpServers": {
    "Math": {
      "command": "uvx",
      "args": ["vibe-math-mcp"]
    }
  }
}

3. If you are developing locally, configure the Math MCP server with a local path reference.

{
  "mcpServers": {
    "Math": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/vibe-math-mcp",
        "run",
        "vibe-math-mcp"
      ]
    }
  }
}

Additional setup and verification

If you use Claude Code or Claude Desktop, you can also add an MCP configuration snippet to your project. For published-package usage, you can insert a standard stdio configuration to ensure the MCP server starts as part of your workflow.

Verify that the server is reachable by listing your configured MCP servers or by querying the available endpoints in your IDE. You should see the Math MCP server available as configured.

Examples of what you can do

• Calculate 15% of 250 using the percentage tool.

• Find the determinant of a 2x2 matrix with matrix_operations and return the result.

• Compute the integral of x^2 from 0 to 1 using integral.

• Evaluate future value of an investment with compound_interest given a present value, rate, and periods.

Notes on usage

Batch execution is available to chain multiple calculations in a single request, enabling complex workflows and reducing round-trips. This is especially useful for financial models, statistical pipelines, or multi-step transformations.

Available tools

calculate

Evaluate mathematical expressions with variable substitution and return the result.

percentage

Perform percentage calculations, including of a value, percentage change, or percentage increase/decrease.

round

Round numbers with options for floor, ceil, truncation, or standard rounding.

convert_units

Convert units such as degrees to radians and vice versa.

array_operations

Perform element-wise operations like add, subtract, multiply, divide, or power on arrays.

array_statistics

Compute descriptive statistics such as mean, median, stddev, min, max, and sum.

array_aggregate

Aggregate arrays with operations like sumproduct, weighted average, and dot product.

array_transform

Transform data via normalization, standardization, scaling, or logarithmic transformation.

statistics

Describe data, compute quartiles, identify outliers, and summarize distributions.

pivot_table

Create pivot tables with customizable aggregations.

correlation

Compute correlation matrices (Pearson, Spearman).

financial_calcs

Time value of money calculations including PV, FV, PMT, IRR, and NPV.

compound_interest

Calculate compound interest with various compounding frequencies.

matrix_operations

Matrix operations such as multiply, inverse, transpose, determinant, and trace.

solve_linear_system

Solve linear systems Ax = b.

matrix_decomposition

Decompose matrices (eigen, SVD, QR, Cholesky, LU).

derivative

Symbolic and numerical differentiation.

integral

Symbolic and numerical integration.

limits_series

Compute limits and series expansions.

batch_execute

Orchestrate multiple tools in a single batch for multi-step workflows.