• MONITOR
  • TRIGGER
  • OPTIMIZE
  • HUMAN

How to manage pre-aggregation jobs with your AI

Connect Cube.dev to your agent to handle data freshness and performance on demand. You can check on running builds or kick off new ones without leaving your chat interface.

Ask AI about this page

Short answer

How can I manage Cube.dev pre-aggregations with AI?

Your agent uses the Cube.dev MCP to check the status of current jobs or start new ones. This lets you maintain data freshness and query speed through simple natural language commands. You'll see the specific outcomes below.

The outcomes.

Control your data freshness.

Your agent handles the heavy lifting of monitoring and triggering builds.

MONITOR

Track job status

The AI checks if your pre-aggregation builds are running, succeeded, or failed. You get immediate visibility into your data pipeline's health.

TRIGGER

Start new builds

When data changes, tell your agent to start a new pre-aggregation job. It handles the request to Cube.dev immediately.

OPTIMIZE

Improve query speed

By managing builds through your agent, you ensure pre-aggregations are ready before users run heavy queries. This keeps your dashboard performance high.

HUMAN

Manual oversight

The AI reports status and triggers actions, but you decide when a build is necessary or if a failure requires manual intervention in the Cube dashboard.

The workflow

What your AI does when a job request arrives.

The agent acts as the bridge between your natural language commands and the Cube.dev API.

  1. Check current status

    The agent first looks at what is currently happening in your Cube environment to see if any jobs are already in progress.

    get_pre_aggregation_job_status

  2. Evaluate the need

    It compares the current job status against your request to determine if a new build is actually required.

    get_pre_aggregation_job_status

  3. Execute the build

    If a new build is needed, the agent sends the command to Cube.dev to start the pre-aggregation process.

    trigger_pre_aggregation_job

  4. Confirm completion

    The agent monitors the new job and reports back to you once the build is finished or if it hits an error.

    get_pre_aggregation_job_status

Try it

Copy these to start.

Use these prompts to begin managing your Cube.dev jobs.

Starting points

These are basic ways to start. Swap out the specific job names or cube names to match your actual setup.

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.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 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"
    }
  }
}
Copy into chat04

  • What is the status of the current pre-aggregation job in Cube.dev?

  • Trigger a new pre-aggregation job for the sales data cube.

  • Check if the pre-aggregation build for the user_activity cube has finished.

  • Are there any failed pre-aggregation jobs I should know about?

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Cube.dev once, then ask.

Just link your Cube.dev instance to Vinkius. Your credentials stay encrypted and secure, and you can immediately start talking to your data through any compatible AI client.

Connect Cube.dev to your AI

FAQ

How this task behaves.

  • 01

    Can the AI delete pre-aggregations?

    No. The AI can only check status or trigger new jobs. It cannot delete existing pre-aggregations.

  • 02

    Can the AI change my Cube schema?

    No. The AI is limited to managing job execution and status. Schema changes must be done manually in your code or Cube dashboard.

  • 03

    Does the AI run the builds itself?

    No. The AI tells Cube.dev to run the builds. The actual computation happens within your Cube environment.

  • 04

    Can I use this with Claude?

    Yes, as long as you are using an MCP-compatible client like Claude Desktop.

  • 05

    Will the AI automatically trigger builds when data changes?

    No. The AI only triggers builds when you specifically ask it to or when you set up a separate automation to prompt the agent.

  • More questions about Cube.dev? The Connector page answers them. See everything the Cube.dev Connector can do

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