Skip to main content
Access your Tableau dashboards, workbooks, and data sources with natural language prompts. Let Gumloop handle the API complexity while you focus on extracting insights and automating your analytics workflows.
Full access to views, workbooks, data sources, and Tableau Pulse metrics so you can retrieve visualizations, export data, and generate AI-powered insights from a single node.

How to Use MCP Nodes

What is Tableau MCP?

Tableau MCP creates an AI-powered node that understands the Tableau REST API for you. Instead of navigating complex endpoints or writing code, simply describe what you want and the node handles the rest. Results come back as structured data that you can pass to other nodes in your workflow.

What Can It Do for You?

  • List and search workbooks, views, and data sources across your Tableau site
  • Export view data as CSV or capture dashboard images as PNG
  • Access Tableau Pulse metrics and generate AI-powered insight bundles
  • Power repeatable workflows by returning structured data that connects to Sheets, Slack, and more

Available Tools

ToolWhat It DoesExample Use
List ViewsList all views on a Tableau site with optional filtering”List views in project ‘Finance’ and return id, name, workbook”
Get View DataRetrieves view data in CSV format”Get data from view view_id and return as CSV”
Get View ImageRetrieves view image as PNG”Get image of view view_id at 1200x800 pixels”
List WorkbooksList all workbooks on a Tableau site with optional filtering”List workbooks owned by ‘john.doe’ and return id, name, project”
Get WorkbookRetrieves workbook details including contained views”Get workbook workbook_id and return name, views, owner”
List DatasourcesList all published data sources on a Tableau site”List datasources in project ‘Sales’ and return id, name, type”
Search ContentSearch across all content types (workbooks, views, datasources, etc.)”Search for ‘revenue’ across all content types and return matching items”
List All Pulse Metric DefinitionsList all Pulse Metric Definitions on the site”List all Pulse metric definitions and return id, name, datasource”
List Pulse Metrics From Definition IdList Pulse Metrics for a specific definition”Get metrics for definition definition_id and return id, specification”
List Pulse Metrics From Metric IdsGet Pulse Metrics by their IDs”Get metrics with ids metric_id_1, metric_id_2 and return details”
List Pulse Metric SubscriptionsList Pulse Metric Subscriptions for current user”List my Pulse metric subscriptions and return metric_id, subscription_id”
Generate Pulse Insight BundleGenerate AI insights bundle for a Pulse Metric”Generate insights for metric metric_id with bundle type ‘detail‘“

How to Use

1

Create Your Tableau MCP Node

Go to your node library, search for Tableau, and click “Create a node with AI”
2

Add Your Prompt

Drag the Tableau MCP node to your canvas and describe the single action you want to perform.
3

Test Your Node

Run the node to see the results. If it works as expected, you’re all set. If you run into issues, check the troubleshooting tips below.
4

Save and Reuse

Once your Tableau MCP node is working, save it to your library for future workflows.

Setting Up Tableau Credentials

Before using the Tableau MCP node, you need to configure your Tableau credentials in Gumloop. Tableau uses Personal Access Tokens (PAT) for authentication.
1

Generate a Personal Access Token in Tableau

  1. Log in to your Tableau Server or Tableau Cloud account
  2. Click on your profile icon and select My Account Settings
  3. Scroll down to the Personal Access Tokens section
  4. Click Create new token
  5. Enter a Token Name (e.g., “Gumloop Integration”)
  6. Click Create
  7. Important: Copy and save both the Token Name and Token Secret immediately. The secret will only be shown once.
2

Gather Your Tableau Server Information

You will need the following information:
  • Server URL: Your Tableau Server URL (e.g., https://10ax.online.tableau.com for Tableau Cloud, or your on-premise server URL)
  • Site Name: The content URL of your site (found in your Tableau URL after /site/). For the default site, this may be empty.
3

Add Credentials in Gumloop

  1. Go to Gumloop Credentials Settings
  2. Click Add Credential
  3. Select Tableau from the list
  4. Enter the following fields:
    • Server URL: Your Tableau Server URL
    • Site Name: Your site’s content URL
    • Personal Access Token Name: The token name you created
    • Personal Access Token Value: The token secret you copied
  5. Click Save
4

Verify Your Connection

Create a simple Tableau MCP node with a prompt like “List workbooks and return the first 5” to verify your credentials are working correctly.
Personal Access Tokens have the same permissions as your Tableau user account. Keep your token secret secure and never share it. If you suspect your token has been compromised, revoke it immediately in your Tableau account settings and create a new one.

Example Prompts

Here are some prompts that work well with Tableau MCP: List Workbooks:
List all workbooks and return id, name, project name, owner name
Get View Data:
Get data from view `view_id` and return as CSV
Search Content:
Search for "sales" across workbooks and views, return id, name, type
List Views with Filtering:
List views where projectName equals "Marketing" and return id, name, workbook name
Get Workbook Details:
Get workbook `workbook_id` and return name, description, views, owner
Export Dashboard Image:
Get image of view `view_id` at 1600x900 pixels
List Pulse Metrics:
List all Pulse metric definitions and return id, name, datasource
Start with a simple prompt like “List workbooks” to confirm your connection works, then add filters and specific fields. Keep prompts focused on a single action for best results.

Troubleshooting

If your Tableau MCP node isn’t working as expected, try these best practices:

Keep Prompts Simple and Specific

  • Good: “List workbooks in project ‘Finance’ and return id, name”
  • Less Efficient: “List all workbooks, filter by project, get their views, export each view as CSV, and combine into a report”
While the complex prompt might work, it’s more efficient to break it into separate nodes. Tableau MCP works best with focused, single-action prompts.

Match What Tableau Can Do

  • Good: “Get data from view view_id and return as CSV”
  • Less Efficient: “Get view data, analyze trends, and create a PowerPoint presentation”
Tableau MCP excels at data retrieval and export. For analysis or presentation creation, connect the output to Ask AI or other nodes in your workflow.

Break Complex Tasks Into Steps

For best results and to avoid timeouts or errors, break large tasks into smaller prompts. Instead of:
List all workbooks, get views from each, export data from every view, and summarize the results
Create a series of standalone nodes:
1

Step 1: List Workbooks

List workbooks and return id, name
2

Step 2: Get Workbook Views

Get workbook workbook_id and return views
3

Step 3: Export View Data

Get data from view view_id and return as CSV
Connect these nodes sequentially in your workflow. The workbook IDs from Step 1 feed into Step 2, and the view IDs from Step 2 feed into Step 3.

Focus on Data Retrieval

Tableau MCP is great at getting information from Tableau. For analysis or content creation, connect it to other nodes. Example:
  • Good prompt: “Get data from view view_id and return as CSV”
  • Less Efficient: “Get view data, calculate year-over-year growth, and draft an executive summary”
Use Ask AI to analyze or summarize results. Keep Tableau prompts focused on retrieving data that other nodes can use.

Troubleshooting Node Creation

In the node creation window, click “Request changes” and ask the AI to add debug logs and verify the API response.
In the node creation window, click “Request changes” and describe what you expected versus what you received.
Verify your Personal Access Token is still valid in your Tableau account settings. Tokens can expire or be revoked. Create a new token if needed.
First click “Fix with Gummie”. If multiple attempts fail, simplify your prompt or contact support.

Need More Help?