mcpField
T
Tableau Mcp

Tableau Mcp

Local
github.com/tableauData & Analytics361Updated today
Repository

About Tableau Mcp

Tableau’s official MCP server provides developer primitives for building AI applications that work with Tableau. It supports a managed hosted endpoint as well as self-hosted and local deployments, with OAuth 2.1 sign-in and per-user Tableau Cloud permissions on the hosted service.

Key features

  1. 01Tools, resources, and prompts
  2. 02Hosted service with OAuth 2.1
  3. 03Tableau Cloud and Tableau Server support
  4. 04Local and standalone executable deployments

Use cases

  1. 01Querying data sources in natural language
  2. 02Finding workbook and view content

MCP support

The server exposes tools, resources, and prompts for MCP clients integrating with Tableau. Its hosted service uses OAuth 2.1 and enforces each user's Tableau Cloud permissions.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "tableau": {
      "command": "npx",
      "args": [
        "-y",
        "@tableau/mcp-server"
      ],
      "env": {
        "SERVER": "<SERVER>",
        "SITE_NAME": "<SITE_NAME>",
        "PAT_NAME": "<PAT_NAME>",
        "PAT_VALUE": "<PAT_VALUE>"
      }
    }
  }
}
Run commandnpx -y @tableau/mcp-server
Environment variables to set
SERVER
SITE_NAME
PAT_NAME
PAT_VALUE

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@tableau/mcp-server

Tools

21

Saved tool list · updated 3w ago

ParameterTypeRequiredDescription
datasourceLuidstringYes

Details

From the project README

Tableau MCP

Tableau Supported

Build and Test

npm

Overview

Tableau MCP is a suite of developer primitives, including tools, resources and prompts, that will make it easier for developers to build AI applications that integrate with Tableau.

Official Documentation

https://tableau.github.io/tableau-mcp/

Getting Started

Hosted Tableau MCP (Recommended for Tableau Cloud)

Tableau MCP is available as a managed service at https://mcp.tableau.com. It uses OAuth 2.1 so every user signs in with their own Tableau Cloud identity, and all existing per-user permissions are enforced automatically.

Point any MCP-compatible client at https://mcp.tableau.com and complete the OAuth sign-in flow when prompted.

See Popular Client Integrations for step-by-step setup instructions for Slack, Claude, ChatGPT, and other common AI clients.

Tableau Server customers and Cloud customers who require self-hosted infrastructure should see the Enterprise Deployment and Self-Hosted Getting Started guides.

Self-Hosted / Local (npx)

The @tableau/mcp-server npm package contains the Web server for Tableau Cloud and Tableau Server. It does not contain the Tableau Desktop authoring server or its data.

This is the quickest way to run Tableau MCP locally. It requires Node.js 22.7.5 or later — no cloning or building needed. Configure your AI tool (MCP client) with:

{
  "mcpServers": {
    "tableau": {
      "command": "npx",
      "args": ["-y", "@tableau/mcp-server@latest"],
      "env": {
        "SERVER": "https://my-tableau-server.com",
        "SITE_NAME": "my_site",
        "PAT_NAME": "my_pat",
        "PAT_VALUE": "pat_value"
      }
    }
  }
}

For Docker, building from source, and other self-hosted options, see the Getting Started guide.

Tableau Desktop Authoring Server

The npm package is Web-only. The Desktop authoring server is a separate build from this repo; see docs/README.desktop.md for the quickstart.

Standalone Binaries (SEA)

The server can be packaged as a Node.js Single Executable Application so it runs without a Node.js install. npm run build:sea -- --variant default builds the Web executable into build/sea/default/<platform>/.

For a published GitHub Release, the upload-binaries workflow is configured to attach one archive per supported release platform. Each archive contains the Web executable (tableau-mcp; .exe on Windows) only. Releases that predate the archives may not have them.

Deploy to Heroku

Deploy to Heroku

Example Prompts to Client

Prompt #1: Querying Data
For the Superstore Datasource, what are the top 5 states with the most sales in 2025?
Response
Based on the Superstore Datasource, here are the top 5 states with the most sales in 2025:
  1. California - $146,388.34
  2. New York - $93,922.99
  3. Washington - $65,539.90
  4. Texas - $43,421.76
  5. Pennsylvania - $42,688.31
California leads by a significant margin with over $146,000 in sales, followed by New York with nearly $94,000. These two states alone account for a substantial portion of the total sales across the top 5 states.
Prompt #2: Content Exploration
Find me the most viewed workbook within the last year.
Response
The most viewed workbook within the last year is:

Global Connections
- Owner: John Doe
- Project: Marketing
- Views in last year: 523
- Last Modified: September 2, 2025
- Workbook ID: 111fb672-9401-4913-a167-17b3d89a3e4f
Prompt #3: Getting a View
Show me the "Economy" view in the "Finances" project.
Response

Economy View Image

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.