How does this MCP ensure my data is accurate?

The MCP uses Cube's semantic layer. Instead of the AI writing its own SQL from scratch, it uses your pre-defined measures and dimensions, ensuring the results match your official business logic.

Can I use this with Cube Cloud?

Yes. If you provide a CUBE_CLOUD_API_KEY, you can use capabilities to list deployments, list environments, and generate metadata tokens.

What AI clients can I use with this MCP?

You can use this MCP with any compatible client, including Claude, Cursor, Windsurf, and VS Code.

Do I need to write SQL myself?

No. You can use natural language to ask questions, and the agent will use capabilities like load_query to get the data. You only need to use SQL capabilities if you specifically want to debug or run raw queries.

Can I manage my pre-aggregations through the AI?

Yes. You can trigger new pre-aggregation build jobs and check their status directly through your agent.

Can I see the exact SQL that Cube generates for a specific query?

Yes. You can use the get_sql capability. By providing the query JSON, the agent will return the generated SQL string, which is perfect for debugging or verifying your data logic.

How do I refresh the data cache or pre-aggregations using the AI?

You can use the trigger_pre_aggregation_job capability. You can specify which cubes or data sources to target, and the agent will initiate the background build process for you.

Is it possible to explore the available measures and dimensions?

Absolutely. Use the get_meta capability to fetch all metadata. This allows the AI to understand what data is available to be queried, including views and segments.

An astronaut overseeing a futuristic data hub

Download PDF Markdown Version

Connect Cube.dev to ChatGPT, Claude and Gemini

Set it up once and it works in ChatGPT, Claude, Cursor, and any other client you love. It all runs safely in the background, so you can relax and focus on what really matters. Turn natural language into precise, governed data queries.

Ask AI about this Connector

Developed, maintained, and hosted by Vinkius.

MCP VERIFIED · PRODUCTION READY · VINKIUS GUARANTEED

Waiting for input…

Works with modern AI clients that support MCP, including ChatGPT, Claude, Cursor, and more.

Your AI does the work. We keep it safe and sound.

No other AI tool gives you this. Your workflow runs in the client you already love, your account stays protected, and you decide what the agent can and cannot do. It is the whole reason to make the switch.

Each Connector lives inside our infrastructure, the same place where your workflow runs. The three things below are already part of it, working quietly in the background so you never have to think about them. Nothing to add, nothing to set up, nothing to worry about.

V8 Isolate01

Each Connector runs inside its own sealed environment.

Each Connector runs inside its own sealed V8 runtime — no access to your files, your system, or other servers. A Connector never sees past its own request.

Kill Switch02

One switch stops every agent. Instantly.

One action deactivates every Connector, revokes every token, and ends every session at once. AI Governance keeps the record.

FinOps per User03

Spending limits that pause instead of surprise.

Every user keeps their own connections and credentials, isolated from everyone else. Anything that would cross your spending limit is held for your approval, not billed.

Three controls, built into the layer your AI connects through.

Curated set · 15 capabilities

We ship only the actions that matter.

Capability set01 / 04

01-04

4 capabilities in this set.

Part of 15 available through Cube.dev.

01
Execute cube SQL
This capability runs a raw SQL query against the Cube SQL API.
02
Generate meta token
This capability creates a JWT for the Metadata API using your CUBE_CLOUD_API_KEY.
03
Get meta
Use this to pull metadata for your cubes and views.
04
Get pre aggregation job status
This capability checks the current status of your pre-aggregation jobs.

Capability set02 / 04

05-08

4 capabilities in this set.

Part of 15 available through Cube.dev.

05
Get SQL
Use this capability to retrieve the generated SQL for a specific Cube query for debugging.
06
List deployments
This capability lists all Cube Cloud deployments if you provide a CUBE_CLOUD_API_KEY.
07
List entities
Use this to see a full list of all available cubes and views.
08
List environments
This capability lists the environments for a specific deployment using your CUBE_CLOUD_API_KEY.

Capability set03 / 04

09-12

4 capabilities in this set.

Part of 15 available through Cube.dev.

09
Load query
This capability executes a Cube query to return aggregated data results.
10
Trigger pre aggregation job
This capability starts a new pre-aggregation build job.
11
Check live
This capability verifies if your Cube deployment is currently live.
12
Check ready
Use this to confirm if your Cube deployment is ready for use.

Capability set04 / 04

13-15

3 capabilities in this set.

Part of 15 available through Cube.dev.

13
Convert query
Use this to turn a standard SQL query into the Cube REST API format.
14
Get entity
This capability retrieves detailed metadata for a specific entity within your data model.
15
List data sources
This capability returns a list of all your configured data sources.

Real work

What teams actually run

Metric Verification

An analytics engineer asks the agent to check if a specific measure matches the expected output in the semantic layer.

Automated Reporting

A product manager asks for a summary of user activity, and the agent pulls the data via the REST API.

SQL Debugging

A developer uses the agent to see why a specific query is returning unexpected results by inspecting the generated SQL.

Cache Management

A data engineer tells the agent to refresh the data for a specific cube to speed up a dashboard.

Automate a task

Put Cube.dev to work on one thing.

Step-by-step automations your AI runs through this Connector: the workflow, the prompts to start with, and the exact capabilities behind each move.

One connector, every AI

Cube.dev works with the most popular AI clients.

These are the most popular clients, each with a step-by-step guide: one link, set up once, with governance and visibility built in. And because everything runs on the MCP standard, the same connection also works in any other compatible client — nothing to rebuild.

Building your own app? The connector is yours to use.

You don't need a client to put Cube.dev to work: the same hosted connection plugs into your own applications and agent code, with the same governance on every request. Build with it, chat with it — one connection for both.

Observed, not estimated

980ms average. Fast in production.

Cube.dev is checked daily against the live service.

Daily averagePeak 1098ms

Oct 3Today

Fastest day: 764ms
Slowest day: 1098ms
14-day trend: Stable+4%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 15 capabilities arrive ready to run.

Preview access · not provider authentication

The vk_preview_* token belongs to Vinkius preview infrastructure. It lets Claude discover and display the capabilities of Cube.dev, so you can see the experience inside your AI.

It does not authenticate your account with Cube.dev. Actions requiring credentials or live account data may not run until you activate the Connector and authorize the service.

Cube.dev Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_umZADPejupUUQHrfZkQT0lvv7MlK4j1JkrhOJOMy/mcp

Claude Desktop

Follow the steps below to connect in seconds.

1In Claude Desktop, open Settings → Connectors.
2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
3Click Add and start a new chat — Cube.dev capabilities are ready to use.

Configuration · claude_desktop_config.jsonCopy

{
  "mcpServers": {
    "cubedev-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_umZADPejupUUQHrfZkQT0lvv7MlK4j1JkrhOJOMy/mcp"
    }
  }
}

Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline

Step-by-step instructions for each client are in the guide. How to connect

Guided setup for Claude? Set up Cube.dev in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work Cube.dev owners hand off.

This MCP is built for technical users who need to bridge the gap between natural language and governed data models.

01
Data Engineers
They use the agent to verify data models and trigger cache refreshes.
02
Analytics Engineers
They use the agent to debug generated SQL and ensure metric consistency.
03
Product Managers
They use the agent to get instant answers to business questions by querying the semantic layer.

FAQ

Questions Cube.dev owners ask.

01
How does this MCP ensure my data is accurate?
The MCP uses Cube's semantic layer. Instead of the AI writing its own SQL from scratch, it uses your pre-defined measures and dimensions, ensuring the results match your official business logic.
02
Can I use this with Cube Cloud?
Yes. If you provide a CUBE_CLOUD_API_KEY, you can use capabilities to list deployments, list environments, and generate metadata tokens.
03
What AI clients can I use with this MCP?
You can use this MCP with any compatible client, including Claude, Cursor, Windsurf, and VS Code.
04
Do I need to write SQL myself?
No. You can use natural language to ask questions, and the agent will use capabilities like load_query to get the data. You only need to use SQL capabilities if you specifically want to debug or run raw queries.
05
Can I manage my pre-aggregations through the AI?
Yes. You can trigger new pre-aggregation build jobs and check their status directly through your agent.
06
Can I see the exact SQL that Cube generates for a specific query?
Yes. You can use the get_sql capability. By providing the query JSON, the agent will return the generated SQL string, which is perfect for debugging or verifying your data logic.
07
How do I refresh the data cache or pre-aggregations using the AI?
You can use the trigger_pre_aggregation_job capability. You can specify which cubes or data sources to target, and the agent will initiate the background build process for you.
08
Is it possible to explore the available measures and dimensions?
Absolutely. Use the get_meta capability to fetch all metadata. This allows the AI to understand what data is available to be queried, including views and segments.

Explore

More in Brain Trust

Connect Cube.dev to Claude, Cursor, ChatGPT & more

Your AI does the work. We keep it safe and sound.

Each Connector runs inside its own sealed environment.

One switch stops every agent. Instantly.

Spending limits that pause instead of surprise.

We ship only the actions that matter.

01-04

Execute cube SQL

Generate meta token

Get meta

Get pre aggregation job status

05-08

Get SQL

List deployments

List entities

List environments

09-12

Load query

Trigger pre aggregation job

Check live

Check ready

13-15

Convert query

Get entity

List data sources

What teams actually run

Metric Verification

Automated Reporting

SQL Debugging

Cache Management

Building your own app? The connector is yours to use.

Cube.dev Connector

Claude Desktop

Data Engineers

Analytics Engineers

Product Managers

How does this MCP ensure my data is accurate?

Can I use this with Cube Cloud?

What AI clients can I use with this MCP?

Do I need to write SQL myself?

Can I manage my pre-aggregations through the AI?

Can I see the exact SQL that Cube generates for a specific query?

How do I refresh the data cache or pre-aggregations using the AI?

Is it possible to explore the available measures and dimensions?

MindsDB (AI Database & Predictors) AI Connector

DataFrame Aggregator Engine AI Connector

data.world AI Connector

DataRobot AI Connector

Suggestions

Data Pipeline Prover AI Connector

Dataiku DSS AI Connector

Hevo Data (ETL & Data Pipeline) AI Connector

Alation AI Connector