mrnh/rigor

Overview

mrnh/rigor is a Python CLI and MCP server that provides verified statistical inference for AI agents. It includes classical and non-parametric hypothesis testing, correlation/regression, effect sizes, power/sample-size calculations, multiple-comparisons correction, a test-recommendation helper, and batch pairwise comparisons.

Best for

Best for
AI developers who need quick statistical verification in Python-based workflows.

Use cases

• Run hypothesis tests from a command line
• Integrate statistical inference into AI agents via MCP
• Perform power analysis and sample size calculations

How to use

Install

pip install rigor-mcp

Tested with

Claude Desktop, Claude Code

Notes

mrnh/rigor is a Python CLI and MCP server that provides verified statistical inference for AI agents. It includes classical and non-parametric hypothesis testing, correlation/regression, effect sizes, power/sample-size calculations, multiple-comparisons correction, a test-recommendation helper, and batch pairwise comparisons.

1 stars on GitHub. Last updated 2026-08-19. Licensed MIT.

Use cases

• Run hypothesis tests from a command line
• Integrate statistical inference into AI agents via MCP
• Perform power analysis and sample size calculations

Pros

• Broad set of statistical tools in one CLI and MCP server
• Includes test-recommendation helper and batch pairwise comparisons
• Supports both classical and non-parametric methods

Cons

• Early-stage project with minimal community adoption (1 star)
• Requires Python environment
• Limited to the listed statistical methods, not a full statistics package

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.