Skip to content
Tableau MCP

Tableau MCP

by Tableau
Released
v4.5.1
Apache-2.0
Repository Docs npm
Featured Databases & Data
tableaubusiness-intelligenceanalyticsdatavisualizationvizqlpulse

Summary

Tableau's own MCP server — query published data sources, explore workbooks and views, and render live vizzes inside an AI client.

Features

  • Run VizQL queries against published Tableau data sources
  • Search and browse workbooks, projects, views and custom views
  • Return a view's data or a rendered image, or draw a live interactive viz in-chat
  • Read and generate Tableau Pulse metric definitions and insight briefs
  • Manage extract refresh tasks, jobs and users, with confirm steps on destructive calls
  • Hosted endpoint with OAuth 2.1 per-user permissions, or self-host with a PAT
  • Separate tool group for driving a running Tableau Desktop instance

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

  1. Open your agent's MCP configuration file.
  2. Merge the snippet below into it, filling in the environment variables with your own values.
  3. Restart the agent — the "Tableau MCP" tools will be available.
{
  "mcpServers": {
    "tableau-mcp": {
      "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"
      }
    }
  }
}

Description

Tableau MCP is Tableau's officially supported MCP server, carrying the "Tableau Supported" badge rather than being a community port. It exposes the Tableau REST, Metadata, VizQL and Pulse APIs as MCP tools, so an assistant can answer questions from governed data instead of a spreadsheet someone pasted into the chat.

What an agent can do with it
  • Query data, not files. query-datasource runs VizQL queries against a published data source, with get-datasource-metadata and list-datasources for discovery. The answers inherit the datasource's governed definitions, so the numbers match what the dashboards show.
  • Explore content. search-content, list-workbooks, list-projects, list-views and get-workbook cover finding things; get-view-data and get-view-image return a view's underlying data or a rendered PNG, with matching tools for custom views.
  • Pulse metrics. Seven tools cover metric definitions, subscriptions, insight bundles and generated insight briefs — the narrative layer, not just the numbers.
  • Prep flows. list-flows, get-flow, list-flow-runs and list-flow-tasks.
  • Administration. Extract refresh tasks (with an explicit confirm step for updates), jobs, users, admin insights, token revocation and a two-step confirm-before-delete for content.
  • Authoring and inline rendering. Workbook upload with a validation step before publish, and render-interactive-viz plus get-embed-token, which draw a live interactive visualisation inside MCP Apps-capable clients.
  • Tableau Desktop. A separate tool group talks to a running Desktop instance: list worksheets and dashboards, read the workbook XML, and apply changes back.
Deployment

The recommended path for Tableau Cloud is the hosted service at https://mcp.tableau.com, which uses OAuth 2.1 so each user signs in with their own Tableau identity and existing per-user permissions apply automatically — no shared service account. Tableau Server customers and anyone needing self-hosted infrastructure run it locally with npx -y @tableau/mcp-server@latest (Node.js 22.7.5+) authenticated by a personal access token, in Docker, or from source. A one-click Heroku deploy template is also provided.

Apache-2.0.

Related MCP Servers

MCP: Meilisearch MCP Server

by Meilisearch

New

Official MCP server for Meilisearch — create and configure indexes, add documents, run hybrid and semantic searches, and manage API keys and tasks through conversation.

Web & SearchDatabases & Data

MCP: Metabase MCP Server

by Metabase

New

Metabase's built-in MCP server — search, query and visualise your BI data through the semantic layer, scoped to the connecting person's existing Metabase permissions.

Databases & DataProductivity
2 views

MCP: Amplitude MCP Server

by Amplitude

New

Amplitude's official hosted MCP server with 50+ tools over analytics, cohorts, experiments, feature flags, taxonomy and session replay, secured by OAuth and existing RBAC roles.

Databases & DataProductivity
1 views

MCP: deja-vu

by Vladislav Shulcz

New

Indexes the coding-agent sessions already sitting on your disk — months of history from before you installed it — and serves them back over MCP. 3.5 GB searched in ~1.5 ms, no LLM, no embeddings.

Developer ToolsDatabases & Data
1 views
Browse all MCP servers →