About tableaugen-mcp
tableaugen-mcp is an MCP server for creating editable Tableau workbooks from CSV files and natural-language requests. It serves users building dashboards with AI clients, and can also profile datasets, preview workbooks, and publish them to Tableau Cloud or Tableau Server.
Key features
- 01Generate .twb and .twbx workbooks
- 02Profile CSV schemas and data
- 03Publish to Tableau Cloud or Server
- 04List dashboard templates
- 05Preview generated workbooks
Use cases
- 01Create a dashboard from a CSV prompt
- 02Inspect a dataset before workbook creation
- 03Publish a generated workbook to Tableau Cloud
MCP support
The server exposes tools for workbook generation, dataset profiling, publishing, templates, and previews to MCP-compatible clients such as Claude Desktop, Cursor, and VS Code.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
5Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| style | string | No | Optional style hint: 'corporate', 'executive', 'analytical', or a custom description. |
| prompt | string | Yes | Natural-language description of the dashboard to generate (e.g. 'quarterly revenue by region with YoY comparison'). |
| csv_path | string | Yes | Absolute path to the source CSV file. |
| output_path | string | Yes | Absolute path where the generated workbook should be saved. Extension determines format (.twb or .twbx). |
Details
From the project READMETabGen MCP Server
Generate production-ready Tableau workbooks from natural language and a CSV.
Official MCP (Model Context Protocol) server for TabGen — the AI-powered Tableau workbook generator. Turn a CSV plus a one-line prompt into a real .twb / .twbx file that opens directly in Tableau Desktop or publishes to Tableau Cloud.
What it does
TabGen writes to Tableau. Point it at a dataset, describe the dashboard you want, and get back an actual editable workbook — not a screenshot, not a suggestion, not a code snippet. The generated file has real data connections, calculated fields, worksheets, dashboard containers, and filter actions wired together.
Use it directly from Claude Desktop, Cursor, VS Code, or any MCP-compatible client.
Installation
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"tableaugen": {
"command": "npx",
"args": ["-y", "tableaugen-mcp"]
}
}
}
Cursor / Windsurf / VS Code
Add to your MCP settings:
{
"tableaugen": {
"command": "npx",
"args": ["-y", "tableaugen-mcp"]
}
}
Direct install
npm install -g tableaugen-mcp
Then run:
tableaugen-mcp
Authentication
None required. TabGen's core generation is free to use.
Tools
| Tool | Description |
|---|---|
generate_workbook | Generate a Tableau .twb / .twbx from a CSV path and a natural-language prompt |
profile_dataset | Inspect a CSV and return schema, data types, cardinality, and null counts |
publish_to_cloud | Publish a generated workbook to Tableau Cloud or Tableau Server |
list_templates | Return available dashboard templates (sales, OEE, finance, marketing, etc.) |
preview_workbook | Return a thumbnail image of a generated workbook for inline preview |
Supported inputs
- CSV files (any dialect, headers auto-detected)
Supported outputs
.twb— Tableau workbook (XML, no data embedded).twbx— Tableau packaged workbook (workbook + extract).hyper— Tableau extract only- Direct publish to Tableau Cloud / Tableau Server via REST API
Example usage
Once installed, ask your AI client:
Generate a quarterly sales dashboard from
/data/orders.csvwith a region filter and a product drill-down. Save it to~/Desktop/sales.twbx.
TabGen will profile the data, plan the layout, and write a valid Tableau workbook to the path you specified.
Links
- Website: tableaugen.com
- Issues: github.com/twilize5/tableaugen-mcp/issues
License
MIT
Comments
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.