Model Context Protocol

Point your assistant
at your warehouse.

Ask Claude or ChatGPT about your business and get the answer from your own data. Add Dataki as a connector, sign in once, and your assistant queries the sources you have connected, saves dashboards and publishes embeds as you. It never sees a database credential.

endpoint streamable http
https://dataki.ai/api/mcp

Nothing to install · no API key to paste · your permissions and spend limits apply · every query logged.

01 Connecting

Add the connector. Sign in. Ask.

Claude and ChatGPT need only the endpoint. They sign you in to Dataki, where you approve what they may do and can change it any time.

  1. 01

    Add a connector in your assistant and give it this URL.

    https://dataki.ai/api/mcp

    In Claude: Add custom connector.

  2. 02

    Sign in when it asks, and approve what it may reach.

    Your assistant gets its own credential, revocable at any time. No config files.

  3. 03

    Ask a question.

    No data connected yet? It hands you a link to connect a source, which takes about a minute, and carries on from there.

Step by step, for the assistant you use:

Under the hood that is OAuth 2.1 with PKCE and dynamic client registration: the client discovers where to send you, you sign in and approve the scopes, and it gets its own revocable credential tied to your account. The server acts only as you: it reaches exactly what you can reach.

Using a client that reads a config file?

Claude Code, Cursor and VS Code read the same endpoint from a file. Create an API key, on any plan, under API keys in the app's account menu and paste one of these.

Claude Code · Cursor mcp.json
{
  "mcpServers": {
    "dataki": {
      "type": "http",
      "url": "https://dataki.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_DATAKI_API_KEY"
      }
    }
  }
}
VS Code .vscode/mcp.json
{
  "servers": {
    "dataki": {
      "type": "http",
      "url": "https://dataki.ai/api/mcp",
      "headers": {
        "Authorization": "Bearer ${input:dataki_key}"
      }
    }
  }
}

02 In the conversation

The answer arrives as a chart you can keep.

In Claude, and in any host that draws MCP Apps, Dataki answers with a live view beside the chat: charts, tables and headline numbers exactly as they appear on a Dataki dashboard. Other hosts get the same answer as text.

Drawn from the warehouse
Each chart carries its query and Dataki runs it, so every figure comes straight from your warehouse.
Checked before you see it
Dataki checks every chart against its data before you see it, and tells the assistant exactly what to fix.
One click to a dashboard
Add to dashboard saves the chart as a live Dataki widget, on a new dashboard or one you already have.
Claude · dataki (MCP) render_ui

How has revenue moved this year, and which categories carry it?

Dataki's view inside Claude: a line of monthly revenue rising from about $169K to $359K over twelve months, and bars of the top six categories led by Outerwear & Coats and Jeans, each with an Add this view to a dashboard button

Dataki in Claude, on Google's public e-commerce sample dataset.

03 Tools

37 tools, in the order you would use them.

An assistant finds the sources, reads what the columns mean, then queries. The rest turn answers into dashboards, embeds and alerts your team keeps.

  • list_sources List data sources

    Start here — the ids every other tool expects.

  • get_data_context Read a source's tables and what they mean

    Tables, columns and types, with metric definitions, joins and column warnings.

  • record_data_context Write back what it learned

    So the next session starts where this one ended.

  • query Run a query

    Read-only SQL across one or more sources, within your spend limits, with a chart and table in hosts that show views.

  • render_ui Show something in the conversation

    Charts, tables and scorecards from live queries, checked before you see them.

  • refresh_view Run a view again

    New dates or filters on a chart already in the conversation.

  • save_dashboard Save a dashboard

    Turn an answer into a live dashboard with its own link.

  • save_widget Add a widget

    Put one chart on a board that already exists.

  • update_widget Change a widget

    Fix a chart's query, title or look where it is.

  • remove_widget Remove a widget

    Take one off a board.

  • list_dashboards List dashboards

    The boards you can add to.

  • add_to_dashboard Choose a dashboard

    Pick the board a chart goes on, from the conversation.

  • get_dashboard Read a dashboard

    What is on a board right now.

  • update_dashboard Update a dashboard

    Its title, theme, pages or who can see it.

  • delete_dashboard Delete a dashboard

    Yours, when it is done.

  • publish_embed Publish an embed

    A public link, and the snippet to put it in a page.

  • watch_metric Watch a metric

    An email when a number crosses the line you set.

  • list_schedules List watches

    Every metric you are watching.

  • delete_schedule Stop a watch

    Turn off a watch you no longer need.

  • upload_file Add a file as a data source

    A CSV or Excel export, queried and charted like any other source.

  • replace_file Replace a file's contents

    Today's export in, the same table names kept: every dashboard on it shows the new rows.

  • create_file_upload Upload a large file

    A signed URL for a file too big to send in the call.

  • list_bigquery_projects List Google Cloud projects

    The projects your Google account can connect, the ones with BigQuery data first.

  • connect_bigquery_project Connect BigQuery

    Add a project's datasets to your team, through a read-only service account.

  • connect_database Connect a database

    PostgreSQL, MySQL or Redshift, tested before it is saved; the password stays encrypted in Dataki.

  • connect_google_sheet Connect a Google Sheet

    A sheet's tabs as tables, from its link.

  • manage_source Look after a source

    Change a database's settings, remove a source, or put back an earlier version of a file.

  • edit_data_context Draft, trace or remove definitions

    Drafts from your tables, checked against the data; who changed each definition and when.

  • generate_dashboard Have Dataki build a dashboard

    A whole board planned and built from your data in a minute or two.

  • share_dashboard Share or copy a dashboard

    Give people or teams access, take it away, or make your own copy.

  • list_teams List teams and their people

    Who is in each team, their roles and the invitations waiting.

  • manage_team Run a team

    Create, rename, leave or delete one; answer an invitation.

  • manage_team_members Manage a team's people

    Invite, remove, change roles, hand the team over.

  • team_usage See or cap a team's usage

    This month's messages against the allowance, and the budget cap.

  • manage_organization Run an organization

    Its teams, people, budget and company sign-in.

  • manage_api_keys Manage keys and embed apps

    Narrower, shorter-lived keys only: never more than this connection holds.

  • account_access Check your access

    Your plan or trial, and which team pays.

04 Verifying

Check it from a terminal first.

Open a session yourself before you add it to an assistant. A valid key answers with the server's name and what it offers. A request with no key gets a 401 that points your client to sign-in.

More about the AI assistant
initialize shell
curl -sX POST https://dataki.ai/api/mcp \
  -H "Authorization: Bearer YOUR_DATAKI_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"1"}}}'
what the agent then sends query
SELECT plan, SUM(mrr) AS mrr
FROM   billing.subscriptions
WHERE  status = 'active'
GROUP BY plan
ORDER BY mrr DESC

Your first month is free.

Connect a source, add the connector, and ask your first question.