ThoughtSpot MCP. Query Metadata and Data Connections from Chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
ThoughtSpot MCP Server lets your AI client talk directly to your ThoughtSpot metadata. You search for specific reports, check user access levels, or list all active data sources—all from a chat window.
Instead of navigating complex UIs, you ask about your organization's BI assets and get the raw information instantly.
What your AI agents can do
List account users
Retrieves a full list of every user account within your ThoughtSpot environment.
List answers
Lists all saved Answers, which are individual charts or tables generated on the platform.
List data connections
Shows you every configured data source connection (like Snowflake and BigQuery) serving ThoughtSpot.
Find specific report types (Liveboards, Answers) by querying metadata using search criteria.
List every configured data source connection the platform uses to power reports and dashboards.
Retrieve lists of all users, user groups, and their corresponding access rights within the ThoughtSpot instance.
Fetch a complete list of existing Liveboards (dashboards) without having to click through category menus.
View all metadata tags used throughout the platform, helping you classify or locate related reports.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
ThoughtSpot MCP Server: 7 Tools for BI Metadata & Governance
These seven tools let your AI client query the underlying structure of ThoughtSpot. You can list everything from user accounts to backend database connections.
019d7612list account users
Retrieves a full list of every user account within your ThoughtSpot environment.
019d7612list answers
Lists all saved Answers, which are individual charts or tables generated on the platform.
019d7612list data connections
Shows you every configured data source connection (like Snowflake and BigQuery) serving ThoughtSpot.
019d7612list liveboards
Provides a list of all existing Liveboards, or main dashboards, in the platform.
019d7612list metadata tags
Retrieves every tag used to classify and categorize your data objects within ThoughtSpot metadata.
019d7612list user groups
Lists all defined user groups and their associated permissions in the system.
019d7612search metadata
Searches for specific metadata objects—Liveboards, Answers, or Logical Tables—by type across your instance.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with ThoughtSpot, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
ThoughtSpot MCP Server lets your AI client talk directly to your ThoughtSpot metadata. You search for specific reports, check user access levels, or list all active data sources—all from a single chat window. Instead of navigating complex UIs, you just ask about your organization's BI assets and get the raw information instantly.
Searching for BI Assets: You use search_metadata to find what you need by querying metadata; it lets you pinpoint specific report types like Liveboards, Answers, or Logical Tables across your entire instance. This tool is how you search through organizational assets when you know the type of data object you're looking for.
Listing Dashboards and Reports: You can get a complete inventory of existing dashboards (Liveboards) by running list_liveboards, which pulls everything without making you click through category menus. For individual charts or tables, use list_answers to list all saved Answers generated on the platform.
Managing User Access: This server lets you manage and view your user directory data. You can pull a full roster of every account in your ThoughtSpot environment using list_account_users. Furthermore, if you need to check who has what rights, list_user_groups shows all defined user groups along with their associated permissions within the system.
Auditing Data Connections: To see where your numbers are coming from, run list_data_connections; this tool provides a list of every configured data source connection—like Snowflake or BigQuery—that's actually powering reports and dashboards on the platform. You know exactly what backend topologies you’re working with.
Identifying Data Classifications: The server also helps you map out your organizational structure using metadata tags. list_metadata_tags retrieves every tag used to classify or categorize all of your data objects within ThoughtSpot metadata, which is helpful if you're trying to locate related reports or group assets by department.
Think about what this means for your workflow: you don’t gotta mess with the dashboard menu system or hunt through multiple tabs. You just ask your agent a question—like 'show me all dashboards tagged 'Finance'' or 'who can see the Q3 revenue report?'—and it spits out the raw data right there, letting you move faster.
This whole setup lets you audit everything from user permissions and data sources to every chart created. You're pulling meta-data on demand, getting a quick read on your entire BI footprint without leaving the chat window.
How ThoughtSpot MCP Works
- 1 Subscribe to this server and provide your ThoughtSpot Cloud Host URL, Username, and Secret Key.
- 2 Start asking your AI agent questions about your BI infrastructure. For example: 'List all active data connections.'
- 3 The agent uses the relevant tool (like
list_data_connections) to pull structured data directly into the chat for immediate review.
The bottom line is you get instant, programmatic access to your BI system's internal structure, bypassing the standard UI.
Who Is ThoughtSpot MCP For?
This server is built for people who need to know what data exists and how it's connected. Data Engineers are tired of manually checking connection strings in a console; Analytics Managers waste time trying to remember which Liveboard belongs to which department; IT Admins spend too much time cross-referencing user permissions across multiple dashboards.
Checks the operational status of backend database connections. Uses list_data_connections to verify if Snowflake or BigQuery links are active before a major migration.
Quickly verifies available reporting assets and organizational structure. Uses list_liveboards and search_metadata when they need to prove the existence of a report that someone claims is missing.
Audits user access and group memberships across the entire organization. Runs list_user_groups or list_account_users to validate compliance without logging into the admin panel.
What Changes When You Connect
- Check data source health instantly. Use
list_data_connectionsto verify if your Snowflake or BigQuery endpoints are online, without having to check the UI settings page. - Find reports by name, not by department. Run
search_metadatato locate a Liveboard ID or Answer chart using keywords across the entire platform's metadata. - Understand who has access. Use
list_user_groupsto see all defined groups and their hierarchy—critical for compliance audits—before checking individual user permissions. - Audit your dashboards in bulk. Running
list_liveboardsgives you a complete inventory of every dashboard available, which is faster than navigating the dashboard menu. - Track data definitions. Use
list_metadata_tagsto see what classification system the organization uses, helping you understand how reports are grouped and governed.
Real-World Use Cases
The Data Engineer needs source verification.
A DE is tasked with confirming if a new reporting dashboard relies on an active AWS Redshift connection. Instead of navigating the data governance section, they ask their agent to run list_data_connections. The system returns a list of all connected sources and their status immediately.
The Analytics Manager needs asset inventory.
An AM is asked to compile a list of every available dashboard for quarterly review. They use the agent's list_liveboards tool, getting an instant list they can copy and paste, saving hours of manual clicking through department portals.
The IT Admin needs group compliance checks.
An IA suspects a user has too much access. They use list_user_groups to see all defined logical groups. They then cross-reference this with list_account_users to pinpoint who belongs where, ensuring the principle of least privilege is enforced.
The Analyst needs a specific chart definition.
An analyst knows a report exists but can't find it. They use search_metadata, specifying the type 'ANSWER' and keywords like 'Q4 Revenue'. The agent returns direct links to the asset's metadata, solving the mystery instantly.
The Tradeoffs
Assuming a UI view is enough.
Trying to guess if data source credentials are correct just by looking at a Liveboard. You might see a dashboard exists but not know if its backend connection is dead or deprecated.
→
Always run list_data_connections first. This tells you the status of every configured data source, giving you operational truth.
Manual user enumeration.
Having to click through multiple 'user groups' and then expand each group to get a full count of users. This is slow and error-prone.
→
Just ask the agent to run list_account_users or list_user_groups. It provides the entire directory listing in one shot.
Over-reliance on visual search.
Searching for a report by remembering its general topic, but not knowing if it's classified as an 'Answer' or a 'Liveboard'.
→
Use search_metadata and specify the asset type you are looking for (e.g., LIVEBOARD, ANSWER) to narrow down your search immediately.
When It Fits, When It Doesn't
You need this server if your primary goal is to audit or inventory the structure of your ThoughtSpot environment without running a query against live data. Use it when you need lists: user names, group memberships, dashboard IDs, or connection status.
Don't use this if you are trying to run an ad-hoc calculation (e.g., 'What was sales last month?'). For that, you should connect your agent directly to the data source layer via a standard BI tool API call. This MCP server is for metadata and asset management only.
If you need to find what reports exist, use list_liveboards or search_metadata. If you need to know who can see those reports, check list_user_groups and list_account_users. It’s a structural audit tool, not an analysis execution tool.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by ThoughtSpot. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 7 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking your BI metadata shouldn't require three different admin tabs.
Today, checking the status of your company's reports is a multi-step chore. You have to navigate to the Admin portal, then check the 'Metadata Tags' section for classifications, and *then* switch over to the Liveboards tab just to see what dashboards are available in that tag category. It’s slow, and you always risk missing an asset.
With this MCP server, you ask your agent one question: 'What reports do we have?' The agent runs `list_liveboards` or `search_metadata`, returning a structured list of everything—the report name, ID, and type—in seconds. You get the full inventory without touching the UI.
ThoughtSpot MCP Server: Audit user access with `list_user_groups`.
Before this tool, figuring out who had rights to a dashboard meant checking the group permissions attached to that specific Liveboard. If you needed to know if 'Marketing' could see anything related to 'Finance,' you had to manually compare two separate groups across multiple dashboards.
Now, you run `list_user_groups`. You get the complete list of every defined group—Administrators, Finance Viewers, Marketing Edit—in a single command. This lets you audit access boundaries quickly and accurately.
Common Questions About ThoughtSpot MCP
How do I find all active data connections using `list_data_connections`? +
You simply ask your agent to run the list_data_connections tool. It returns a list of every configured source, including Snowflake and BigQuery details, so you know what feeds the platform.
What is the difference between `list_answers` and `list_liveboards`? +
Liveboards are full dashboards—collections of multiple charts. Answers are individual assets; they're single charts or tables that can be used in a dashboard, which you find using list_answers.
Can I search for reports by type with `search_metadata`? +
Yes. The search_metadata tool lets you filter results specifically by asset type like LIVEBOARD, ANSWER, or LOGICAL_TABLE, making your search highly accurate.
`list_account_users` shows me everyone in the company, right? +
Correct. Running list_account_users gives you a full enumeration of all user accounts registered within the ThoughtSpot instance for auditing purposes.
What happens if my user lacks permissions when I use `search_metadata`? +
The tool respects your current access level. If you don't have permission to view a specific asset, the server simply won't return it in the results list. You get clean error handling instead of a crash.
Does running `search_metadata` let me filter by tags using `list_metadata_tags`? +
Yes, you can combine these calls. While list_metadata_tags shows available tags, you pass those tag values into the search parameters for search_metadata. This lets you narrow down results quickly.
What does an empty response mean when I run `list_liveboards` or `list_answers`? +
It means the asset simply doesn't exist in your ThoughtSpot environment. The tool isn't failing; it's just reporting that zero liveboards or answers matched your query criteria.
How does the MCP Server authenticate using my ThoughtSpot Cloud Host URL, Username, and Secret Key? +
You provide those credentials during setup. The server uses them to establish a direct, authenticated connection to your specific ThoughtSpot instance. This keeps your data isolated from other environments.
Where do I get my ThoughtSpot Secret Key? +
You can find or generate your Secret Key in the ThoughtSpot instances under the Develop menu, inside the Security settings (specifically Trusted Authentication). Note that you need admin tracking rights or developer privileges to enable this type of access.
Can the AI modify my dashboards? +
No. The integration exclusively lists and searches metadata elements (Liveboards, Data connections, Answers, Users). It will not modify, overwrite, or delete any configurations or data objects.
What types of metadata can I search for? +
Liveboards (dashboards), Answers (individual charts and tables), Logical Tables, data source connections, users, user groups, and organizational tags.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Spotify Listening History Parser
Parse your Spotify data export and discover your top artists, tracks, and total listening hours. Turn your AI into your personal music analyst local.
Storm Glass Marine & Weather
Universal marine and weather intelligence — get global wave, tide, and bio data via AI.
Google Air Quality
Universal air quality intelligence — get Google-powered AQI, pollutants, and health advice via AI.
You might also like
HERE Mobility
AI transit planning: discover routes, schedules, and stations for public transport via agents.
KanbanZone
Manage projects visually with multi-board Kanban views, WIP limits, and process metrics that streamline team delivery.
U.S. EIA Energy Data
Equip your AI agent to access official U.S. energy statistics, track electricity generation, and monitor fuel prices via the EIA API.