Dataki for Cursor

Query your database from Cursor

Let Cursor's agent query your real data while you code. Dataki's hosted MCP server lets it read the schema, run read-only queries on the databases you have connected and save what it finds as a dashboard, with your connection string safe in Dataki.

  1. Cursor
  2. Dataki MCP server
  3. Your database, read-only
Signs in with
A Dataki API key, created for you when you click install.
Works in
Cursor's chat and agent.
You need
Cursor, and a Dataki account with at least one data source.
Server URL
https://dataki.ai/api/mcp
Connect Cursor See the steps

01Set it up

3 steps, a couple of minutes

Prefer buttons? Dataki's connect page walks you through the same steps and creates a key wherever one is needed.

  1. 01

    Add Dataki to Cursor from the connect page

    Open the Cursor steps on Dataki's connect page and click Add to Cursor. It creates a key for Cursor and opens Cursor with Dataki ready to install.

  2. 02

    In Cursor, click Install

  3. 03

    Ask it something in Cursor's chat

    Start with What data can you see in Dataki?.

Prefer to add it by hand?

Create a key on Dataki's connect page and put it in ~/.cursor/mcp.json. The file holds the key, so keep it in your home directory, outside any repository.

~/.cursor/mcp.json 
{
  "mcpServers": {
    "dataki": {
      "type": "http",
      "url": "https://dataki.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_DATAKI_KEY"
      }
    }
  }
}

02Ask it

What to ask in Cursor's chat

The assistant finds your sources, learns what the columns mean and writes the SQL. Dataki runs it read-only, and every answer shows its query.

Which tables in the billing database hold subscriptions, and how do they join to customers?

Answered from get_data_context: the tables, columns and types, plus any metric definitions and join paths your team has recorded for that source.

How many users signed up each day since Monday's release, and how does that compare with the week before?

The assistant writes the SQL and Dataki runs it read-only. The query comes back with the answer, ready to check.

Are there any orders without a matching customer? Show me the ten most recent.

A production data check in one sentence, with no SQL client to open and no connection string in your editor.

Put a chart of daily signups for the last 90 days on a dashboard called Growth.

save_dashboard turns it into a live Dataki dashboard with its own link. It refreshes on its own, and anyone you share it with can open it, no editor needed.

03What it can reach

Any source you connect to Dataki

Cursor reaches every source your Dataki account can open, and one question can draw on several of them.

04Worth knowing

What Cursor gets, and what it never does

Treat the key like a password
It acts as you and reaches what you can reach in Dataki. Every key has daily limits on queries and bytes scanned, every query it runs is logged under its name, and you can revoke it from Dataki at any time. Keep it out of your repositories.
No connection string in your config
Your editor holds one revocable key. The database credentials stay encrypted in Dataki, and every query runs with a read-only role.
Your app's data, kept apart from your code
Dataki answers from the database: signups, orders, events. It never sees your code.

More on the server itself, and every tool it offers, on the MCP server page. How credentials, keys and the query log work: security and governance.

Dataki and Cursor

How do I connect Cursor to a PostgreSQL database with MCP?
Connect the database to Dataki with a read-only user, then add Dataki's hosted MCP server to Cursor: one click on Dataki's connect page, or an entry in ~/.cursor/mcp.json with a Dataki API key. Cursor's agent can then read the schema and run read-only queries while you code.
Does Cursor get my database password?
No. Your database credentials stay encrypted in Dataki and never reach Cursor or any model. Cursor calls Dataki's MCP server, Dataki runs the SQL with a read-only role, and only the results reach the conversation.
Can Cursor change or delete data through Dataki?
No. Dataki is read-only end to end: any statement that changes data is refused before it runs, and the connection itself uses a role that cannot write. Cursor can create and edit Dataki dashboards, which live in Dataki, never in your database.
Which databases can Cursor query through Dataki?
Any source connected to Dataki: Google BigQuery, PostgreSQL, MySQL, Amazon Redshift, Supabase, Google Sheets and uploaded CSV or Excel files, including hosted databases such as Neon, Amazon RDS, Cloud SQL and PlanetScale. Google Analytics 4, Firebase, Search Console, Firestore and Stripe work through their exports to BigQuery. One question can draw on several sources.

Using another assistant?

Connect Cursor. Ask it something.

Connect a source to Dataki, add the server to Cursor, and your first answer is one question away. Anything worth keeping becomes a live dashboard with its own link.

Connect Cursor Talk to us