Tableau

What you can do

Connect this agent connector to let your agent:

• Browse workbooks and views — List, search, and retrieve detailed metadata for workbooks, views (sheets and dashboards), and data sources published on a Tableau site
• Export visualizations — Download dashboards as PNGs, PDF documents, or Excel crosstab files; download full workbooks as .twbx files
• Query underlying data — Export view summary data as CSV, or run structured queries against published data sources using the VizQL Data Service API (Tableau Cloud 2024.1+)
• Monitor jobs — Poll background job status to track completion of long-running operations
• Manage the site — Create and update projects; add and remove users; create, add, and remove groups

Authentication

Tableau uses Personal Access Token (PAT) authentication. You store your PAT credentials in Scalekit once, and Scalekit calls tableau_auth_signin to obtain a session token, then refreshes it automatically before every tool call. Your agent code never handles tokens directly — Scalekit injects the current session token as the X-Tableau-Auth header on every request.

How it works:

1. Store your PAT name, PAT secret, domain, and site content URL in a Scalekit connected account
2. Before each tool call, Scalekit checks if the session token is still valid (with a 5-minute buffer)
3. If the token has expired or is about to expire, Scalekit signs in automatically using the stored PAT credentials via tableau_auth_signin
4. The fresh session token is injected as X-Tableau-Auth — your code does nothing
5. After sign-in, the site ID (site LUID) is stored automatically in the connected account — you do not pass site_id to tool calls. Token lifetime is 120 minutes for Tableau Cloud and 240 minutes for Tableau Server.

Set up the connector

Connect your Tableau Cloud or Tableau Server site to Scalekit so your agent can browse workbooks, query views, export dashboards, and manage users.

Scalekit handles session token management automatically. You store your Personal Access Token (PAT) credentials once, and Scalekit signs in and refreshes the session token before it expires — your code never calls the sign-in endpoint directly.

1. Create a Personal Access Token in Tableau

   A Personal Access Token (PAT) is used by Scalekit to sign in on your behalf and keep the session alive automatically.

   • Sign in to your Tableau site.
   • Click your avatar in the top-right corner → My Account Settings.
   • Scroll to the Personal Access Tokens section.
   • Click + Create new token, give it a name (e.g., scalekit-agent), and click Create.
   • Copy both the Token Name and Token Secret — the secret is shown only once.

   

   Find your site content URL

   Your site content URL is the identifier in your Tableau browser URL. For https://prod-in-a.online.tableau.com/#/site/mycompany-1234567, the site content URL is mycompany-1234567. Leave it blank if you use the Default site.

2. Create a connection in Scalekit

   • In Scalekit dashboard, go to Agent Auth → Create Connection.
   • Search for Tableau and click Create.
   • Note the Connection name — use this as connection_name in your code (e.g., tableau).
   • Click Save.

   

3. Add a connected account

   A connected account links a user in your system to their Tableau PAT credentials. Scalekit uses these to sign in and refresh the session automatically.

   Via dashboard (for testing)

   • Open the connection → Connected Accounts tab → Add account.
   • Fill in:
     • Your User’s ID — any identifier for this user (e.g., user_123)
     • Server Domain — your Tableau hostname without https:// (e.g., prod-in-a.online.tableau.com)
     • PAT Name — the token name from step 1
     • PAT Secret — the token secret from step 1
     • Site Content URL — the site identifier from your Tableau URL (leave blank for the Default site)
   • Click Save.

   

   Via API (for production)

   Node.js

   await scalekit.actions.upsertConnectedAccount({
     connectionName: 'tableau',
     identifier: 'user_123',
     credentials: {
       domain: 'prod-in-a.online.tableau.com',
       pat_name: 'scalekit-agent',
       pat_secret: process.env.TABLEAU_PAT_SECRET,
       site_content_url: 'mycompany-1234567', // omit for Default site
     },
   });

   Python

   scalekit_client.actions.upsert_connected_account(
       connection_name="tableau",
       identifier="user_123",
       credentials={
           "domain": "prod-in-a.online.tableau.com",
           "pat_name": "scalekit-agent",
           "pat_secret": os.getenv("TABLEAU_PAT_SECRET"),
           "site_content_url": "mycompany-1234567",  # omit for Default site
       },
   )

   Automatic token refresh

   Scalekit refreshes the session token before every tool call using the stored PAT credentials. Tokens are renewed automatically when they are within 5 minutes of expiry — you do not need to manage refresh logic in your code.

Code examples

Once a connected account is set up with PAT credentials, your agent can call Tableau tools through Scalekit. The session token is managed and injected automatically as the X-Tableau-Auth header — your code never handles it directly.

The site ID (site LUID) is resolved automatically from the connected account after sign-in. You do not pass site_id to tool calls. For proxy API calls that require a site ID in the URL path, call tableau_session_get once to retrieve it.

Get the connected account

Node.js

import { ScalekitClient } from '@scalekit-sdk/node';
import 'dotenv/config';

const scalekit = new ScalekitClient(
  process.env.SCALEKIT_ENV_URL,
  process.env.SCALEKIT_CLIENT_ID,
  process.env.SCALEKIT_CLIENT_SECRET
);
const actions = scalekit.actions;

const connectionName = 'tableau';
const identifier = 'user_123';

const { connectedAccount } = await actions.getConnectedAccount({
  connectionName,
  identifier,
});

Python

from scalekit.client import ScalekitClient
import os

scalekit_client = ScalekitClient(
    client_id=os.getenv("SCALEKIT_CLIENT_ID"),
    client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
    env_url=os.getenv("SCALEKIT_ENV_URL"),
)
actions = scalekit_client.actions

connection_name = "tableau"
identifier = "user_123"

connected_account = actions.get_connected_account(
    connection_name=connection_name, identifier=identifier
).connected_account

Proxy API calls

Use the Scalekit proxy to call any Tableau REST API endpoint directly. Binary downloads (PNG, PDF, Excel, .twbx, .tdsx) must use the proxy — use tableau_session_get to retrieve the site ID for the URL path:

Node.js

// Get site ID once (needed for proxy URL construction)
const session = await actions.executeTool({
  toolName: 'tableau_session_get',
  connectedAccountId: connectedAccount.id,
  toolInput: {},
});
const siteId = session.session.site.id;

// Export a view as PNG
const imageBytes = await actions.request({
  connectionName,
  identifier,
  path: `/api/3.28/sites/${siteId}/views/${viewId}/image`,
  method: 'GET',
  queryParams: { resolution: 'high' },
});

// Export a view as PDF
const pdfBytes = await actions.request({
  connectionName,
  identifier,
  path: `/api/3.28/sites/${siteId}/views/${viewId}/pdf`,
  method: 'GET',
  queryParams: { type: 'a4', orientation: 'landscape' },
});

// Download a workbook (.twbx)
const workbookBytes = await actions.request({
  connectionName,
  identifier,
  path: `/api/3.28/sites/${siteId}/workbooks/${workbookId}/content`,
  method: 'GET',
});

// Download a data source (.tdsx)
const datasourceBytes = await actions.request({
  connectionName,
  identifier,
  path: `/api/3.28/sites/${siteId}/datasources/${datasourceId}/content`,
  method: 'GET',
});

Python

# Get site ID once (needed for proxy URL construction)
session = actions.execute_tool(
    tool_name="tableau_session_get",
    connected_account_id=connected_account.id,
    tool_input={},
)
site_id = session.data["session"]["site"]["id"]

# Export a view as PNG
image_response = actions.request(
    connection_name=connection_name,
    identifier=identifier,
    path=f"/api/3.28/sites/{site_id}/views/{view_id}/image",
    method="GET",
    query_params={"resolution": "high"},
)
with open("dashboard.png", "wb") as f:
    f.write(image_response.content)

# Export a view as PDF
pdf_response = actions.request(
    connection_name=connection_name,
    identifier=identifier,
    path=f"/api/3.28/sites/{site_id}/views/{view_id}/pdf",
    method="GET",
    query_params={"type": "a4", "orientation": "landscape"},
)
with open("dashboard.pdf", "wb") as f:
    f.write(pdf_response.content)

# Download a workbook (.twbx)
workbook_response = actions.request(
    connection_name=connection_name,
    identifier=identifier,
    path=f"/api/3.28/sites/{site_id}/workbooks/{workbook_id}/content",
    method="GET",
)
with open("workbook.twbx", "wb") as f:
    f.write(workbook_response.content)

# Download a data source (.tdsx)
datasource_response = actions.request(
    connection_name=connection_name,
    identifier=identifier,
    path=f"/api/3.28/sites/{site_id}/datasources/{datasource_id}/content",
    method="GET",
)
with open("datasource.tdsx", "wb") as f:
    f.write(datasource_response.content)

No auth header needed

The Scalekit proxy automatically injects the X-Tableau-Auth session token header. You only provide the path and method.

Use Scalekit tools

Browse workbooks and views

Node.js

// List all workbooks on the site
const workbooks = await actions.executeTool({
  toolName: 'tableau_workbooks_list',
  connectedAccountId: connectedAccount.id,
  toolInput: {},
});
// workbooks.workbooks.workbook[] — each has id, name, contentUrl, project

// Search for a workbook by name
const found = await actions.executeTool({
  toolName: 'tableau_workbook_search',
  connectedAccountId: connectedAccount.id,
  toolInput: { name: 'SalesReport' },
});

// List all views within a workbook
const workbookId = workbooks.workbooks.workbook[0].id;
const views = await actions.executeTool({
  toolName: 'tableau_workbook_views_list',
  connectedAccountId: connectedAccount.id,
  toolInput: { workbook_id: workbookId },
});
// views.views.view[] — each has id, name, contentUrl

Python

# List all workbooks on the site
workbooks = actions.execute_tool(
    tool_name="tableau_workbooks_list",
    connected_account_id=connected_account.id,
    tool_input={},
)
# workbooks["workbooks"]["workbook"] — each has id, name, contentUrl, project

# Search for a workbook by name
found = actions.execute_tool(
    tool_name="tableau_workbook_search",
    connected_account_id=connected_account.id,
    tool_input={"name": "SalesReport"},
)

# List all views within a workbook
workbook_id = workbooks["workbooks"]["workbook"][0]["id"]
views = actions.execute_tool(
    tool_name="tableau_workbook_views_list",
    connected_account_id=connected_account.id,
    tool_input={"workbook_id": workbook_id},
)
# views["views"]["view"] — each has id, name, contentUrl

Sign out

Call tableau_auth_signout to invalidate the session token when the agent session ends:

Node.js

await actions.executeTool({
  toolName: 'tableau_auth_signout',
  connectedAccountId: connectedAccount.id,
  toolInput: {},
});
// The stored session token is now invalid — Scalekit will refresh on next call

Python

actions.execute_tool(
    tool_name="tableau_auth_signout",
    connected_account_id=connected_account.id,
    tool_input={},
)
# The stored session token is now invalid — Scalekit will refresh on next call

Getting resource IDs

Most Tableau tools require one or more resource LUIDs. The site ID is resolved automatically by Scalekit after sign-in — you do not pass it to tool calls. Always fetch other IDs from the API — never guess or hard-code them.

Resource	Tool to get ID	Field in response
Workbook ID	tableau_workbooks_list or tableau_workbook_search	workbooks.workbook[].id
View ID	tableau_views_list or tableau_workbook_views_list	views.view[].id
Data Source ID	tableau_datasources_list	datasources.datasource[].id
Project ID	tableau_projects_list	projects.project[].id
User ID	tableau_users_list	users.user[].id
Group ID	tableau_groups_list	groups.group[].id
Job ID	tableau_job_get (from background job operations)	job.id
Site ID (proxy only)	tableau_session_get	session.site.id

Recommended start sequence for any agent session:

1. tableau_workbooks_list           → discover workbooks
2. tableau_workbook_views_list      → discover views within a workbook
3. tableau_datasources_list         → discover data sources

Tool list

tableau_auth_signout Sign out of Tableau, invalidating the current session token. Call this at the end of an agent session. Scalekit will obtain a fresh token automatically on the next tool call. 0 params ▾

Sign out of Tableau, invalidating the current session token. Call this at the end of an agent session. Scalekit will obtain a fresh token automatically on the next tool call.

tableau_session_get Returns information about the current authenticated session, including the site name, site content URL, and the authenticated user. Useful for confirming which site the agent is connected to. 0 params ▾

Returns information about the current authenticated session, including the site name, site content URL, and the authenticated user. Useful for confirming which site the agent is connected to.

tableau_site_get Retrieve information about a Tableau site: name, content URL, storage quota, user quota, and status. Optionally include usage statistics. 1 param ▾

Retrieve information about a Tableau site: name, content URL, storage quota, user quota, and status. Optionally include usage statistics.

Name Type Required Description

include_usage_statistics boolean optional Set to `true` to include storage and user count statistics.

tableau_workbooks_list List published workbooks on a Tableau site. Supports filtering (e.g., `name:eq:SalesReport`, `ownerName:eq:jane`), sorting (`name:asc`, `updatedAt:desc`), and pagination. 4 params ▾

List published workbooks on a Tableau site. Supports filtering (e.g., `name:eq:SalesReport`, `ownerName:eq:jane`), sorting (`name:asc`, `updatedAt:desc`), and pagination.

Name Type Required Description

filter string optional Filter expression, e.g. `name:eq:SalesReport` or `ownerName:eq:jane`.

sort string optional Sort expression, e.g. `name:asc` or `updatedAt:desc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_workbook_search Search for workbooks on a Tableau site by exact name. Returns workbooks whose name matches the search term. 3 params ▾

Search for workbooks on a Tableau site by exact name. Returns workbooks whose name matches the search term.

Name Type Required Description

name string required The workbook name to search for (exact match).

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_workbook_get Retrieve detailed information about a specific workbook: name, owner, project, tags, views, and data connections. Optionally include view count statistics. 2 params ▾

Retrieve detailed information about a specific workbook: name, owner, project, tags, views, and data connections. Optionally include view count statistics.

Name Type Required Description

workbook_id string required Workbook LUID. Get it from `tableau_workbooks_list` → `workbooks.workbook[].id`.

include_usage_statistics boolean optional Set to `true` to include view count and high-water-mark statistics.

tableau_workbook_delete Permanently delete a workbook and all of its views from the Tableau site. This action cannot be undone. 1 param ▾

Permanently delete a workbook and all of its views from the Tableau site. This action cannot be undone.

Name Type Required Description

workbook_id string required Workbook LUID. Get it from `tableau_workbooks_list`. WARNING: This is permanent.

tableau_workbook_connections_list List the data connections used by a workbook: connection type, server address, username, and whether the connection is embedded. 1 param ▾

List the data connections used by a workbook: connection type, server address, username, and whether the connection is embedded.

Name Type Required Description

workbook_id string required Workbook LUID. Get it from `tableau_workbooks_list`.

tableau_views_list List all views (sheets and dashboards) across the entire site. Supports filtering, sorting, and pagination. Use `tableau_workbook_views_list` to scope to a single workbook. 5 params ▾

List all views (sheets and dashboards) across the entire site. Supports filtering, sorting, and pagination. Use `tableau_workbook_views_list` to scope to a single workbook.

Name Type Required Description

filter string optional Filter expression, e.g. `name:eq:SalesDashboard`.

sort string optional Sort expression, e.g. `name:asc` or `viewCount:desc`.

include_usage_statistics boolean optional Set to `true` to include view count statistics.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_workbook_views_list List all views (sheets and dashboards) within a specific workbook. Returns each view's LUID, name, content URL, and owner. 5 params ▾

List all views (sheets and dashboards) within a specific workbook. Returns each view's LUID, name, content URL, and owner.

Name Type Required Description

workbook_id string required Workbook LUID. Get it from `tableau_workbooks_list`.

include_usage_statistics boolean optional Set to `true` to include view count for each view.

filter string optional Filter expression, e.g. `name:eq:Overview`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page.

tableau_view_get Retrieve detailed information about a specific view: name, owner, workbook, content URL, tags, and creation date. 2 params ▾

Retrieve detailed information about a specific view: name, owner, workbook, content URL, tags, and creation date.

Name Type Required Description

view_id string required View LUID. Get it from `tableau_views_list` or `tableau_workbook_views_list` → `views.view[].id`.

include_usage_statistics boolean optional Set to `true` to include total view count.

tableau_datasources_list List published data sources on a Tableau site. Supports filtering (e.g., `name:eq:SalesData`, `type:eq:excel`), sorting, and pagination. 4 params ▾

List published data sources on a Tableau site. Supports filtering (e.g., `name:eq:SalesData`, `type:eq:excel`), sorting, and pagination.

Name Type Required Description

filter string optional Filter expression, e.g. `name:eq:SalesData` or `type:eq:excel`.

sort string optional Sort expression, e.g. `name:asc` or `updatedAt:desc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_datasource_get Retrieve detailed information about a specific published data source: name, type, owner, project, tags, and connection details. 1 param ▾

Retrieve detailed information about a specific published data source: name, type, owner, project, tags, and connection details.

Name Type Required Description

datasource_id string required Data source LUID. Get it from `tableau_datasources_list` → `datasources.datasource[].id`.

tableau_datasource_delete Permanently delete a published data source from the Tableau site. This action cannot be undone and will break any workbooks that depend on this data source. 1 param ▾

Permanently delete a published data source from the Tableau site. This action cannot be undone and will break any workbooks that depend on this data source.

Name Type Required Description

datasource_id string required Data source LUID. Get it from `tableau_datasources_list`. WARNING: This is permanent.

tableau_projects_list List projects on a Tableau site. Projects organize workbooks and data sources. Supports filtering (e.g., `name:eq:Marketing`), sorting, and pagination. 4 params ▾

List projects on a Tableau site. Projects organize workbooks and data sources. Supports filtering (e.g., `name:eq:Marketing`), sorting, and pagination.

Name Type Required Description

filter string optional Filter expression, e.g. `name:eq:Marketing`.

sort string optional Sort expression, e.g. `name:asc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_project_create Create a new project on a Tableau site. Optionally nest it under a parent project and set content permission behavior. 4 params ▾

Create a new project on a Tableau site. Optionally nest it under a parent project and set content permission behavior.

Name Type Required Description

name string required Display name for the new project.

description string optional Optional description.

parent_project_id string optional Parent project LUID to create a sub-project. Get it from `tableau_projects_list`.

content_permissions string optional `ManagedByOwner` (default) or `LockedToProject`.

tableau_project_update Update a project's name, description, parent project, or content permission behavior. 5 params ▾

Update a project's name, description, parent project, or content permission behavior.

Name Type Required Description

project_id string required Project LUID. Get it from `tableau_projects_list` → `projects.project[].id`.

name string optional New display name.

description string optional New description.

parent_project_id string optional New parent project LUID to move the project.

content_permissions string optional `ManagedByOwner` or `LockedToProject`.

tableau_project_delete Permanently delete a project from the Tableau site. Content within the project is moved to the default project (not deleted). This action cannot be undone. 1 param ▾

Permanently delete a project from the Tableau site. Content within the project is moved to the default project (not deleted). This action cannot be undone.

Name Type Required Description

project_id string required Project LUID. Get it from `tableau_projects_list`. WARNING: This is permanent.

tableau_users_list List users on a Tableau site. Supports filtering (e.g., `siteRole:eq:SiteAdministratorCreator`), sorting (`name:asc`, `lastLogin:desc`), and pagination. 4 params ▾

List users on a Tableau site. Supports filtering (e.g., `siteRole:eq:SiteAdministratorCreator`), sorting (`name:asc`, `lastLogin:desc`), and pagination.

Name Type Required Description

filter string optional Filter expression, e.g. `siteRole:eq:Viewer` or `name:eq:jane`.

sort string optional Sort expression, e.g. `name:asc` or `lastLogin:desc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_user_get Retrieve information about a specific user: name, email, site role, last login, and authentication type. 1 param ▾

Retrieve information about a specific user: name, email, site role, last login, and authentication type.

Name Type Required Description

user_id string required User LUID. Get it from `tableau_users_list` → `users.user[].id`.

tableau_user_add_to_site Add a user to the Tableau site with a specified role. If the user account does not exist, it is created. The `site_role` field controls what the user can do. 3 params ▾

Add a user to the Tableau site with a specified role. If the user account does not exist, it is created. The `site_role` field controls what the user can do.

Name Type Required Description

name string required Username or email address of the user to add.

site_role string required Role to assign: `SiteAdministratorCreator`, `SiteAdministratorExplorer`, `Creator`, `ExplorerCanPublish`, `Explorer`, `Viewer`, or `Unlicensed`.

auth_setting string optional Authentication type: `ServerDefault`, `SAML`, or `OpenIDConnect`.

tableau_user_remove_from_site Remove a user from the Tableau site. The user's content (workbooks, data sources) is transferred to the site admin. The user account itself is not deleted from the server. 1 param ▾

Remove a user from the Tableau site. The user's content (workbooks, data sources) is transferred to the site admin. The user account itself is not deleted from the server.

Name Type Required Description

user_id string required User LUID. Get it from `tableau_users_list`.

tableau_groups_list List groups on a Tableau site. Groups simplify permission management — you assign permissions once to a group and they apply to all members. 4 params ▾

List groups on a Tableau site. Groups simplify permission management — you assign permissions once to a group and they apply to all members.

Name Type Required Description

filter string optional Filter expression, e.g. `name:eq:Analytics`.

sort string optional Sort expression, e.g. `name:asc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_group_create Create a new local group on a Tableau site. Optionally set a minimum site role for all group members. 2 params ▾

Create a new local group on a Tableau site. Optionally set a minimum site role for all group members.

Name Type Required Description

name string required Display name for the new group.

minimum_site_role string optional Minimum site role for members: `Viewer`, `Explorer`, `Creator`, etc.

tableau_group_add_user Add a user to a group on a Tableau site. The user must already be a site member. Use this to manage group-based permissions. 2 params ▾

Add a user to a group on a Tableau site. The user must already be a site member. Use this to manage group-based permissions.

Name Type Required Description

group_id string required Group LUID. Get it from `tableau_groups_list` → `groups.group[].id`.

user_id string required User LUID. Get it from `tableau_users_list` → `users.user[].id`.

tableau_group_remove_user Remove a user from a group. The user remains a site member — only group membership is changed. 2 params ▾

Remove a user from a group. The user remains a site member — only group membership is changed.

Name Type Required Description

group_id string required Group LUID. Get it from `tableau_groups_list`.

user_id string required User LUID. Get it from `tableau_users_list`.

tableau_jobs_list List background jobs on a Tableau site. Jobs include extract refreshes, data source imports, and workbook publishes. Filter by status: `InProgress`, `Success`, `Failed`, or `Cancelled`. 4 params ▾

List background jobs on a Tableau site. Jobs include extract refreshes, data source imports, and workbook publishes. Filter by status: `InProgress`, `Success`, `Failed`, or `Cancelled`.

Name Type Required Description

filter string optional Filter expression, e.g. `status:eq:Failed`.

sort string optional Sort expression, e.g. `createdAt:desc`.

page_number integer optional Page number (starts at 1).

page_size integer optional Items per page (max 1000).

tableau_job_get Retrieve the current status and details of a background job: type, status (`InProgress`, `Success`, `Failed`, `Cancelled`), progress percentage, and error details if failed. Use this to poll after triggering a refresh. 1 param ▾

Retrieve the current status and details of a background job: type, status (`InProgress`, `Success`, `Failed`, `Cancelled`), progress percentage, and error details if failed. Use this to poll after triggering a refresh.

Name Type Required Description

job_id string required Job LUID returned from async operations like workbook or data source refreshes → `job.id`.

tableau_job_cancel Cancel a background job that is currently queued or in progress. Already completed, failed, or cancelled jobs cannot be cancelled. 1 param ▾

Cancel a background job that is currently queued or in progress. Already completed, failed, or cancelled jobs cannot be cancelled.

Name Type Required Description

job_id string required Job LUID. Get it from `tableau_jobs_list` or from a refresh response. Only queued/in-progress jobs can be cancelled.