Skip to content
MCPSTAMP

Live dataNo token

A/B test sample size calculator MCP server

Exact sample sizes, test durations and significance tests for A/B tests, including Fisher exact tests.

Free. No key, no sign-in, no paid tier. 5 tools.

Connector URL

https://ab-test-sample-size-calculator-mcp.lipmichal.workers.dev/mcp

Paste it as a custom connector in Claude, or add it to Claude Code, Cursor or VS Code. Setup for each is below.

What it knows that a model does not

A model answers from training data. This server reads the source when you ask.

  • Exact normal quantiles and the full sample size formula. A rounded z-value moves the answer by thousands of users.
  • Why calculators disagree. It shows the pooled formula (as R power.prop.test) next to the formula Evan Miller publishes.
  • Exact Fisher-Freeman-Halton p-values for larger tables, and how much daily peeking inflates false positives.

Sources Checked against R reference examples, scipy and statsmodels, and published sequential testing tables.

Tools (5)

Read from the live server with tools/list on 2026-11-02.

ToolWhat it does
calculate_sample_size A/B test sample size calculator. Exact sample size per variant for an A/B test on a conversion rate (two independent proportions): baseline rate, minimum detectable effect (relative or absolute), alpha, power, one or two-sided, several variants with optional Bonferroni, and unequal allocation.
estimate_test_duration A/B test duration estimator. How many days an A/B test must run: from a known sample size per variant, or from baseline rate and MDE (computed as in calculate_sample_size), plus daily visitors and the share of traffic in the test.
test_ab_significance A/B test significance calculator. Is an A/B test result statistically significant?
fisher_exact_test Fisher exact and Fisher-Freeman-Halton test. Exact p-value for a contingency table of counts. 2x2: Fisher exact test (two-sided, greater or less, as R fisher.test).
check_peeking_risk A/B test peeking and sequential testing risk. What peeking does to an A/B test: the real false positive rate if you check a fixed-horizon test K times and stop at the first p < alpha (for example about 14% at 5 looks with alpha 0.05), and the Pocock and O'Brien-Fleming group...

Connect it

One URL, no token. Pick your app.

Claude.ai and Claude Desktop

Open Settings, then Connectors, then Add custom connector. Paste the URL. Leave the OAuth fields empty. Save.

https://ab-test-sample-size-calculator-mcp.lipmichal.workers.dev/mcp

Claude Code

Run this in your terminal. Then run claude mcp list to check it.

claude mcp add --transport http ab-test-sample-size-calculator https://ab-test-sample-size-calculator-mcp.lipmichal.workers.dev/mcp

Cursor

Merge this into ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project).

{
  "mcpServers": {
    "ab-test-sample-size-calculator": {
      "url": "https://ab-test-sample-size-calculator-mcp.lipmichal.workers.dev/mcp"
    }
  }
}

VS Code

Put this in .vscode/mcp.json. The top key is servers, not mcpServers.

{
  "servers": {
    "ab-test-sample-size-calculator": {
      "type": "http",
      "url": "https://ab-test-sample-size-calculator-mcp.lipmichal.workers.dev/mcp"
    }
  }
}

Things to ask

Paste one into a new chat after you connect.

Questions

Is the a/B test sample size calculator MCP server free?
Yes. The hosted URL is free and needs no token, key or account. The source is open on GitHub.
Why not just ask the model?
A model answers from its training data. That data stops at a cutoff date, and long sums can drift. This server reads its data or rules each time you ask, so the answer is current and exact.
Where does the data come from?
Checked against R reference examples, scipy and statsmodels, and published sequential testing tables.
Which apps can use it?
Any MCP client that accepts a remote server URL: Claude.ai and Claude Desktop custom connectors, Claude Code, Cursor, VS Code and others. You can also run it locally with npx.

More free live-data servers

All free servers ยท Back-office servers on McpStamp