Metabase connector for AI agents

Tools your analytics agent reaches for on Metabase, scoped per user.

CALL ANY TOOL

77 tools for business intelligence in Metabase: run saved questions and SQL queries, create and update cards, dashboards, and collections, and read database, table, and field metadata.

metabase_action_execute

Execute action

Run an action, which writes to the connected database (insert, update, delete, or custom SQL). Returns the action's result, such as rows-affected or created-row. Get the parameter IDs from metabase_action_get first. This changes data in the source database.

Parameters

Name

Type

Required

Description

action_id

integer

Required

ID of the action to run, from metabase_actions_list.

parameters

object

Optional

Parameter values keyed by parameter ID (from metabase_action_get). Example: {"id": 17, "status": "shipped"}.

metabase_action_get

Get action

metabase_actions_list

List actions

metabase_card_copy

Copy card

metabase_card_create

Create card

Create a saved question, model, or metric (card) from a query. Returns the new card, including its id. Use this to save a query for reuse; use metabase_native_query_run to run SQL without saving it.

Parameters

Name

Type

Required

Description

dataset_query

object

Required

The query, as a Metabase query object. Metabase treats this as opaque, so the safest way is to copy the dataset_query of an existing card (metabase_card_get with legacy_mbql=true). A SQL query looks like {"database": 1, "type": "native", "native": {"query": "SELECT * FROM orders LIMIT 10"}}; a table query looks like {"database": 1, "type": "query", "query": {"source-table": 12}}.

display

string

Required

How to visualize the results: table, bar, line, area, row, pie, scalar, smartscalar, combo, pivot, funnel, map, scatter, progress, gauge, waterfall, or object.

name

string

Required

Name of the card.

cache_ttl

integer

Optional

Cache duration multiplier for query results, in hours. Leave empty to use the instance default.

collection_id

integer

Optional

ID of the collection to put the card in. Leave empty for the root collection ('Our analytics'). Find IDs with metabase_collections_list.

collection_position

integer

Optional

Pin position in the collection (1 = first pinned slot). Leave empty to keep the item unpinned.

dashboard_id

integer

Optional

Save the question directly inside this dashboard (a dashboard question) instead of a collection. Leave empty to save it to a collection.

description

string

Optional

Description of the card.

parameters

array

Optional

Parameter (filter) definitions as an array of objects. Each needs an id, name, slug, and type such as category, date/single, or number/=. Example: [{"id": "a1b2c3d4", "name": "Category", "slug": "category", "type": "category"}].

type

string

Optional

Kind of card: question (default), model, or metric. One of: `question`, `model`, `metric`.

visualization_settings

object

Optional

Visualization settings object for the chart (for example column formatting or graph dimensions). Leave empty to use defaults ({}).

metabase_card_delete

Delete card

metabase_card_query

Query card

metabase_card_update

Update card

Update a card's name, description, query, visualization, collection, or archived state. Returns the updated card. Use archived=true to move a card to the trash (recoverable); use metabase_card_delete to delete it permanently.

Parameters

Name

Type

Required

Description

card_id

integer

Required

ID of the card (a saved question, model, or metric). Find it with metabase_search or metabase_cards_list.

archived

boolean

Optional

Set true to move the card to the trash, or false to restore it.

cache_ttl

integer

Optional

Cache duration multiplier for query results, in hours. Leave empty to use the instance default.

collection_id

integer

Optional

Move the card to this collection ID.

collection_position

integer

Optional

Pin position in the collection (1 = first pinned slot). Leave empty to keep the item unpinned.

dashboard_id

integer

Optional

Move the question into this dashboard (makes it a dashboard question).

dataset_query

object

Optional

description

string

Optional

New description.

display

string

Optional

New visualization type: table, bar, line, area, row, pie, scalar, smartscalar, combo, pivot, funnel, map, scatter, progress, gauge, waterfall, or object.

name

string

Optional

New name.

parameters

array

Optional

type

string

Optional

Change the kind of card: question, model, or metric. One of: `question`, `model`, `metric`.

visualization_settings

object

Optional

Visualization settings object for the chart (for example column formatting or graph dimensions). Leave empty to use defaults ({}).

metabase_cards_list

List cards

metabase_collection_get

Get collection

metabase_collections_list

List collections

metabase_dashboard_create

Create dashboard

metabase_dashboard_delete

Delete dashboard

metabase_dashboard_query_metadata_get

Get dashboard query metadata

metabase_dashboard_update

Update dashboard

metabase_database_rescan_values

Database rescan values

metabase_database_schemas_list

List database schemas

metabase_database_sync_schema

Database sync schema

metabase_field_summary_get

Get field summary

metabase_native_query_run

Run native query

metabase_notification_send

Send notification

metabase_notifications_list

List notifications

metabase_search

metabase_segment_get

Get segment

metabase_segment_update

Update segment

metabase_segments_list

List segments

metabase_snippet_create

Create snippet

metabase_table_update

Update table

metabase_tables_list

List tables

metabase_timeline_event_get

Get timeline event

For more tools, view docs.

Build your Agent

Same auth pattern across LangChain, OpenAI, Anthropic, and Google ADK.

import { ScalekitClient } from "@scalekit-sdk/node";
import { createReactAgent } from "@langchain/langgraph/prebuilt";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

// Metabase tools scoped to this user
const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["metabase"], toolNames: [
    "metabase_search",
    "metabase_card_query",
    "metabase_native_query_run"] },
  pageSize: 100,
});

const agent = createReactAgent({ llm, tools });
await agent.invoke({ messages: [{ role: "user", content: "Run the Weekly Signups question and compare it with last month" }] });

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const openai = new OpenAI();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["metabase"] }, pageSize: 100,
});

const res = await openai.chat.completions.create({
  model: "gpt-5",
  messages: [{ role: "user", content: "Run the Weekly Signups question and compare it with last month" }],
  tools,
});

// Execute the tool call with the user's vaulted Metabase API key
await sk.tools.executeTool(res.choices[0].message.tool_calls[0], "user_123");

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);
const anthropic = new Anthropic();

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["metabase"] }, pageSize: 100,
});

const msg = await anthropic.messages.create({
  model: "claude-sonnet-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Run the Weekly Signups question and compare it with last month" }],
  tools,
});

// Tool call runs with the user's vaulted Metabase API key
await sk.tools.executeTool(msg.content, "user_123");

import { Agent } from "@google/adk/agents";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(env.SCALEKIT_ENV_URL, env.SCALEKIT_CLIENT_ID, env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["metabase"] }, pageSize: 100,
});

const agent = new Agent({
  name: "analytics_agent",
  model: "gemini-2.5-pro",
  instruction: "Metabase tools scoped to this user",
  tools,
});

await agent.run("Run the Weekly Signups question and compare it with last month");

Try these prompts

Copy any prompt into your agent. Each maps directly to a Metabase tool. Click to copy, paste into your agent, done.

Questions and queries

Copy the prompt

Copied

Run the Weekly Signups saved question.

Copy the prompt

Copied

Run a SQL query counting orders by status in the analytics database.

Copy the prompt

Copied

Search Metabase for anything about churn.

Dashboards and collections

Copy the prompt

Copied

Create a dashboard called Q4 Growth in the Marketing collection.

Copy the prompt

Copied

List the most viewed dashboards and questions.

Copy the prompt

Copied

Copy the Sales Overview dashboard for the EMEA team.

Data model

Copy the prompt

Copied

List the tables in the public schema of the warehouse database.

Copy the prompt

Copied

Sync the schema so Metabase picks up the new events table.

Copy the prompt

Copied

Add a launch event to the Product Launches timeline for March 4.

SEE HOW AUTH WORKS

Each user connects their own Metabase API key once, and Scalekit sends it with every call. Keys stay vaulted, every call is scope checked, and every action is logged.

Authorize

Your user connects

Metabase

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

Store

Their

Metabase

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

Resolve

When your agent calls a

Metabase

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

Audit

Every

Metabase

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other analytics and BI connectors.

GTM and RevOps Teams

Revenue forecast commentary

Pulls open pipeline from Salesforce and HubSpot, calculates coverage against quota, flags at-risk stages, posts commentary to Slack, and logs every snapshot to Google Sheets.

GTM and RevOps Teams

Competitive intelligence briefing agent

Scans Gong calls for competitor mentions, matches each one to its Notion battlecard, and DMs every affected rep a single Slack digest per cycle. Every call runs as the PMM who owns the briefing, never a shared bot.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

01.

Shared tokens break per-user analytics

A shared Metabase API key looks fine in a demo. In production every query and dashboard change looks like one service account, and you cannot tell which user triggered it. Scalekit resolves the credential of the actual user who triggered the agent, never a shared bot.

// shared key
audit → bot_service_account

// scalekit
audit → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

Metabase today. Ten connectors tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access Metabase as the user or through a shared key?

As the user. Each user connects their own Metabase API key once, and Scalekit sends it with every call. Audit logs attribute every query and dashboard change to that user, not a shared service account.

Where is the Metabase API key stored?

In Scalekit's managed AES-256 token vault, namespaced per tenant. To rotate it, the user reconnects with a new API key. Revocation is a single dashboard action. Credentials never appear in prompts, logs, or LLM context.

Can I limit what the agent does in Metabase?

Yes. Filter by tool name in listScopedTools to expose only what you want, for example the read and query tools without metabase_action_execute, which writes to the connected database, or metabase_card_delete and metabase_dashboard_delete, which cannot be undone. Scalekit also enforces scope checks before every API call.

What happens when a user revokes Metabase access?

The connection is invalidated on the next tool call. Subsequent requests for that user fail closed with a clear error. Other users in the tenant remain unaffected. The event is logged for audit.

Can the agent run SQL against my databases through Metabase?

Yes. metabase_native_query_run runs a SQL query against a connected database without saving it, and metabase_query_run runs any Metabase query object. Queries run with the access of the Metabase user the API key acts as, which metabase_user_current_get returns.