Crosstabs MCP Server

Inspect a worked result before sharing data

Our customer-churn example includes downloadable aggregate counts, statistical results, interpretation, and limitations. The counts are illustrative, not a customer case study. Use it to see what a reviewable result looks like; the hosted demo is a separate synthetic example, not a promise of identical output or independent certification.

Check the published validation scope and preserve sample bases, assumptions, and warnings in any AI-generated explanation.

Connect to the free hosted MCP

This is the public product endpoint. No Crosstabs account or API key is required. Client availability can still depend on the AI app, version, and plan.

https://mcp.crosstabs.com/mcp

1. Call get_started with your client name, then run_demo_analysis. This path creates no workspace and uploads no file.
2. For your own files, review the privacy policy and hosted data flow. Only after informed consent, call create_workspace with consentToUpload: true.
3. Open the returned private upload page, transfer a supported file, and give the resulting uploadId to the AI. The workspace token is a password; never publish it.
4. Ask the AI to analyze your survey with analyze_survey, then explain the supported insights with table citations, sample bases, missing responses, and limitations. Confirm ambiguous columns and weight meaning before analysis. Poll get_job no more frequently than every two seconds, then download work before the 24-hour expiry.

Survey/report files: 5 MiB each. Project files: 8 MiB each. Cumulative uploads: 10 MiB per workspace. Shared storage and job quotas also apply; see hosted capacity limits. Your chosen AI provider has its own handling of tool inputs and results.

Open the temporary workspace utility. The hosted service includes task shortcuts, detailed project operations, and aggregate evidence tools. Deterministic output still requires appropriate assumptions and human interpretation.

Need files to stay on your device?

Use the public endpoint for the simplest experience. Install the local package when policy forbids hosted uploads, when you need the complete standalone statistics catalog, or when a durable local workflow matters more than zero setup.

codex plugin marketplace add barangaroo/crosstabs-codex-plugin --ref 857a93b32dd22f7e655ec30e3c628f70ba9a14d5
codex plugin add crosstabs@crosstabs

pip install crosstabs==1.4.0

# Complete project workflow
crosstabs-headless

# Focused statistical tools
crosstabs

claude mcp add crosstabs-headless -- crosstabs-headless
claude mcp add crosstabs-statistics -- crosstabs

{
  "mcpServers": {
    "crosstabs-headless": {
      "command": "crosstabs-headless"
    },
    "crosstabs-statistics": {
      "command": "crosstabs"
    }
  }
}

Verified package; separate source and directory gates

The published crosstabs==1.4.0 passed fresh public-index checks of both local stdio servers: 35 headless tools and 40 statistics-server registrations (39 analytical tools plus runtime status). Both runtime-status calls confirm the executing package version. The exact active Registry 1.4.0 record was verified separately; historical 1.2.2 evidence remains distinct. The plugin source and parity contract is separate evidence from package publication. Native-host review and directory approval remain pending.

The browser worker and local MCP enter the same HeadlessCrosstabsService and share the TabBookProject schema, initial-analysis policy, deterministic statistical kernels, project revision, audit, replay, and undo orchestration. Cross-surface golden tests remain release gates rather than blanket certification. Local persistence is bounded to 8 MiB per serialized project and 32 MiB per database, including audit, undo and replay state; parser import ceilings are not a persistence guarantee. The plugin does not reproduce browser interaction state or optional local connector authorization. The hosted MCP also exposes the project engine through consented temporary uploads and background jobs. Its public surface now includes simpler task-level entry points while remaining distinct from the complete standalone Python statistical tool set.

Local package capabilities

Example prompt: “Create a project from this survey, profile cardinality, define age and region banners, run an unweighted tab book, export an editable PPTX with audit evidence. Preserve any weight column as a candidate and do not execute it without typed design authority.”

Source and license

The Python 3.10+ package is declared under the MIT license and is distributed on PyPI as a wheel and downloadable source archive. The development repository is currently private; public GitHub access is not claimed here. The separately packaged plugin source repository is public. Report package or plugin issues to support@crosstabs.com. Core methods share documented conventions with the web calculators, while the package also includes additional methods with their own assumptions and output contracts. See the methods and current review status before relying on results for high-stakes work.

Prefer plain HTTP?

The free JSON API exposes chi-square, Fisher's exact, odds ratios, and 20+ measures without installing anything.

Frequently asked questions

What is the crosstabs MCP server?

Crosstabs provides one free hosted survey-analysis MCP endpoint at https://mcp.crosstabs.com/mcp. Connect a compatible Streamable HTTP client without a Crosstabs account or API key and start with the upload-free synthetic demo. Private file workflows require explicit upload consent and a secret, temporary workspace token. An optional local package is available when files must stay on your device.

How do I try Crosstabs without uploading a file?

Connect the hosted endpoint, call get_started and get_runtime_status, then run_demo_analysis. This demonstration uses fixed synthetic counts, creates no workspace, and uploads no respondent data. Ask your AI to explain the result, assumptions, warnings, and next action. Client support may depend on your AI app, version, and plan.

Is there an official Crosstabs plugin for Codex?

Crosstabs plugin 0.4.0 is available from a verified public GitHub source revision and pins crosstabs 1.4.0. Source installation is separate from native-host review and directory approval; no directory submission, approval or listing is claimed.

Does the Codex plugin have full Crosstabs workspace parity?

No. Package 1.4.0 covers the local project workflow: projects, imports, filters, typed survey designs, tab books, waves, schema repair, editable report packs, export, and audit history. Manual open-end coding remains a browser-workspace feature, and browser interaction state is not reproduced by the plugin.

Why use an MCP server instead of letting the AI compute statistics itself?

Language models can make arithmetic mistakes. The MCP server executes numerical routines instead; each method still has its own assumptions, and outputs distinguish exact, asymptotic, and simulated inference.

Does the MCP server send my data anywhere?

The local servers do not upload files to Crosstabs. The hosted MCP does receive consented respondent files and temporary projects. Hosted access expires 24 hours after workspace creation; active cleanup and provider backup retention are separate. Your AI client may also receive tool inputs and results under its own policies.