list_workbooksLABSCO SUMMARY
The numbers on a dashboard become something you can quote.
Getting a figure out of Tableau normally means opening the view, exporting, and finding the download. Pulling the CSV directly means the assistant can answer from the published data, and the image export covers the times a picture is what you actually wanted.
What it is
A Python server connecting an MCP client to Tableau Server or Tableau Cloud. It browses workbooks, views and data sources, extracts the data behind a view, exports a dashboard as a PNG, downloads a whole data source, and searches across the site's content.
What you get
- Workbooks, views and published data sources, listed
- The data behind any view, as CSV
- A dashboard exported as a PNG image
- A whole data source downloaded, with its raw rows
- Search across the site's content, for finding a report by what is in it
Requirements
Python 3.8 or newer and access to Tableau Server or Tableau Cloud. What you can reach follows the account you connect with — administrative privileges are needed for the administrative side. Installed from the repository with pip install -r requirements.txt.
Setup effort
Build from source — git clone https://github.com/hetpatel-11/Tableau-MCP.git
USE CASES
01
Finding the report someone half-described
Search the site's content, then list the views inside the workbook that matches.
02
Quoting a number from a dashboard
Pull the view's data as CSV and read the value, instead of reading it off a screenshot.
03
Putting a dashboard into a document
Export the view as a PNG and drop it straight into the write-up.
04
Working out where a figure comes from
List the published data sources behind the workbook before trusting the chart on top of them.
05
Doing analysis Tableau will not do for you
Download the data source in full and take the raw rows somewhere else.
15 tools
Listed in project docs. Your client may show a different list.
list_viewsget_view_dataget_view_imagesearch_contentlist_datasourcesdownload_datasourcelist_workbook_permissionslist_all_userslist_all_groupsget_user_permissionslist_projects_permissionsget_site_infoaudit_workbook_usagelist_user_groupsBEFORE YOU INSTALL
What to know first
Tools you get
15 tools named in the project's documentation. Eight tools, the permission, user, group, project and site tools, need administrative privileges. Nothing publishes or edits Tableau content.
Requires
Python 3.8+
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
TERMINAL
git clone https://github.com/hetpatel-11/Tableau-MCP.git
cd Tableau-MCP
python -m venv tableau_mcp_env
source tableau_mcp_env/bin/activate # On Windows: tableau_mcp_env\Scripts\activate
pip install -r requirements.txt
cp env.example .envIn .env, set your Tableau Server or Tableau Cloud URL (TABLEAU_SERVER_URL) and site (TABLEAU_SITE_ID), and either a Personal Access Token (TABLEAU_TOKEN_NAME and TABLEAU_TOKEN_VALUE, recommended) or TABLEAU_USERNAME and TABLEAU_PASSWORD
TERMINAL
chmod +x start_clean.shMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"tableau-server": {
"command": "<YOUR_PATH_TO_TABLEAU_MCP>/start_clean.sh",
"args": []
}
}
}
