$ loading_
A read-only MCP server for table discovery, profiling, and statistical tests.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Statistical Testing MCP Server" yet — see the docs or source repo.
List the currently discoverable tables and briefly describe their fields or analyzable scope.
A list of accessible tables with basic structure or scope notes for each.
Create a deterministic profile for the orders table, including row count, key field distributions, and suggested columns for later statistical testing.
A stable, reproducible table overview with column statistics and candidate fields for testing.
Compare the mean difference between treatment and control using Welch's t-test; if it is a conversion-rate difference, use a two-proportion z-test and explain the result.
The output includes the test used, test statistic, significance result, and a brief interpretation.
Data analysts or researchers can use it to run Welch's t-tests or two-proportion z-tests on two groups to assess whether mean or conversion-rate differences are significant. It fits read-only analysis workflows and avoids modifying source data.
When working with an unfamiliar dataset, users can start with bounded table discovery and deterministic profiling to understand available tables, field distributions, and which columns are suitable for later statistical testing.
Developers can connect this MCP server to an AI assistant so the model can discover tables, inspect data, and use well-maintained statistical libraries for common tests in a read-only setup.
It is a read-only MCP server that provides bounded table discovery, deterministic data profiling, and statistical testing. It explicitly supports Welch's t-test and the two-proportion z-test.
Based on the description, it is a read-only service. That means it is intended for discovery, analysis, and testing rather than writing or changing data.
The description mentions SQLite, pandas, SciPy, and statsmodels. More specific installation steps or runtime requirements are not provided in the given material.
Let AI call R and Python statistical functions without writing code.
Run statistical analysis, probability calculations, and data processing through natural language.
Run tests programmatically, inspect results, and get test strategy recommendations.
Use 65 deterministic tools to reliably test MCP protocols across transports.
Analyze MCP tool security risks, detect malicious behavior, and provide risk scores.
Provides math, scientific, conversion, and statistical calculations.