tableau
Tableau's official MCP Server. Helping Agents see and understand data.
claude mcp add --transport stdio tableau-tableau-mcp npx -y @tableau/mcp-server@latest \ --env SERVER="https://my-tableau-server.com" \ --env PAT_NAME="my_pat" \ --env PAT_VALUE="pat_value" \ --env SITE_NAME="my_site"
How to use
Tableau MCP provides a set of developer primitives and prompts to help you build AI apps that integrate with Tableau data and workflows. The server package exposes a configurable MCP endpoint that can be consumed by various MCP clients, enabling capabilities such as querying Tableau data sources, exploring content, and retrieving views or workbooks through structured prompts. By using the included prompts and utilities, you can craft natural-language requests that Tableau can fulfill, with responses formatted for easy display in your client apps. The to-be-installed server is run via npm/npx and will read the environmental variables you provide to connect to your Tableau Server or Tableau Cloud instance.
To use the server, point your MCP client at the configured server endpoint and supply the required credentials via environment variables (SERVER, SITE_NAME, PAT_NAME, PAT_VALUE). The toolkit handles authentication and provides prompts for common Tableau tasks like querying data, exploring workbooks, and retrieving specific views. You can customize prompts or extend capabilities by updating the server configuration or using the published prompts to suit your data and governance needs.
How to install
Prerequisites:
- Node.js 22.7.5 or newer
- npm or corepack available in your environment
- Access to a Tableau Server or Tableau Cloud and a personal access token (PAT)
Installation steps:
- Ensure Node.js is installed:
node -v
# should be >= 22.7.5
- Install or run the MCP server via npx as shown in the Quick Start:
# Using the npm package name from the README example
npx -y @tableau/mcp-server@latest
- Configure environment variables for your Tableau connection:
export SERVER=https://my-tableau-server.com
export SITE_NAME=my_site
export PAT_NAME=my_pat
export PAT_VALUE=pat_value
- Start the MCP server (the exact command may depend on your setup; the example uses npx to run the server descriptor):
# If using a local dev script or a package-provided start script, run it here
# Example placeholder if a start script is provided by the package
npm run start
- Verify the server is accessible by your MCP client and begin issuing prompts such as querying data or exploring workbooks.
Notes:
- If you deploy to environments like Heroku or other hosting platforms, ensure the same environment variables are set in those environments.
- Keep your PAT secure and rotate per your organization’s security practices.
Additional notes
Tips and common considerations:
- The SERVER variable must point to your Tableau Server or Tableau Cloud instance reachable from the MCP host.
- SITE_NAME should match the site you intend to access within Tableau.
- PAT_NAME and PAT_VALUE must be a valid personal access token with appropriate permissions for the requested operations.
- When operating in production, consider limiting access to the MCP endpoints and logging detailed prompts for auditing.
- If you upgrade the mcp-server package, re-check the environment variable names or required permissions as new capabilities are added.
- Some MCP clients may cache configuration; ensure client restarts pick up updated env vars and server settings.
Related MCP Servers
mcp-graphql
Model Context Protocol server for GraphQL
ghost
A Model Context Protocol (MCP) server for interacting with Ghost CMS through LLM interfaces like Claude. Allow you to control your Ghost blog by simply asking Claude etc.
any-script
An MCP server that exposes arbitrary CLI tools and shell scripts as MCP Tools
filesystem
A Model Context Protocol (MCP) server for platform-agnostic file capabilities, including advanced search/replace and directory tree traversal
unplugin
A unified plugin for developers integrating MCP servers into modern JavaScript build tools, including Webpack, Rollup, Vite, and more.
prometheus
A Model Context Protocol (MCP) server implementation that provides AI agents with programmatic access to Prometheus metrics via a unified interface.