# Apache Superset MCP for AI Agents AI Agent Connect

> Apache Superset MCP connects your AI agent to your BI platform. It allows your agent to pull data from dashboards, inspect datasets, and run live SQL Lab queries. Instead of clicking through menus to find a metric, you can just ask your agent to find it, analyze the underlying logic, or pull out a specific slice of data for a report.

## Overview
- **Category:** brain-trust
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_gwYe1NYA9B8gFp8lQ7hoG8Zq04EdrBNzR1eq1ypu/ai-agent-connect
- **Tags:** data-visualization, sql-lab, dashboarding, data-exploration, business-insights

## Description

The Apache Superset MCP connects your AI agent directly to your BI platform. Imagine you need to know why a specific sales metric dropped last week. Usually, you'd have to log into your BI tool, find the right dashboard, hunt for the specific chart, and then dig into the SQL to see what's actually happening. With this connector, you just ask your agent to do the heavy lifting. It can scan your dashboards to find the right one, look at the charts to see what's being visualized, and then jump straight into the database to run a query that explains the discrepancy. It turns your BI platform into a searchable knowledge base where you can grab data for reports or audit your analytics infrastructure without ever leaving your chat window. It's a huge time saver for anyone who spends their day digging through layers of nested menus. You can find this and thousands of other connectors in the Vinkius catalog to get your agent connected to the rest of your stack.

## Tools

### get_chart_details
Pull the specific details and logic for a single chart. This shows you exactly how a metric is being calculated.

### list_charts
See a list of every chart currently available in your Superset instance. Use this to browse your available visualizations.

### get_dashboard_details
Get the specific configuration and metadata for a dashboard. This helps you see how a dashboard is structured.

### list_dashboards
Find all the dashboards in your BI portal to see what's available. This is great for auditing your reporting surfaces.

### list_databases
See all the different data sources your Superset instance is connected to. It helps you verify your backend connections.

### list_datasets
View the full inventory of datasets ready for analysis. This lets you see what data is available to query.

### execute_sql_query
Run a raw SQL statement against a specific database ID in SQL Lab. Use this to pull custom data that isn't in a dashboard.

## Prompt Examples

**Prompt:** 
```
Find our 'Monthly Active Users' dashboard and tell me what charts are on it.
```

**Response:** 
```
I've located the **Monthly Active Users** dashboard. It contains the following charts:

*   **Active Users by Region** (Map visualization)
*   **New Signups vs. Retained Users** (Grouped Bar Chart)
*   **Churn Rate Trend** (Line Chart)
*   **Device Breakdown** (Pie Chart)
```

**Prompt:** 
```
List all the databases connected to our Superset instance.
```

**Response:** 
```
Here are the databases currently connected to your Superset instance:

| Database Name | ID | Connection Type |
| :--- | :--- | :--- |
| **SalesDB** | 12 | PostgreSQL |
| **Marketing_Analytics** | 45 | Snowflake |
| **Web_Logs** | 8 | ClickHouse |
```

**Prompt:** 
```
Run a SQL query on our SalesDB to find the top 3 products by revenue this month.
```

**Response:** 
```
Executing query on **SalesDB**...

Here are the top 3 products by revenue for the current month:

1.  **Electronics Pro** - $1,240,000
2.  **Home Office Suite** - $890,500
3.  **Smart Watch Gen 5** - $720,000
```

## Capabilities

### Pull chart data from dashboards
Extract specific numbers and metrics from your existing visualizations.

### Find the right dashboard
Search your entire BI portal to locate the correct reporting surface.

### Inspect chart logic
See the exact metrics and underlying logic used in any specific chart.

### Run raw SQL queries
Execute pure SQL statements against your connected databases via SQL Lab.

### Audit your data sources
Inventory all connected databases and available datasets in one place.

## Use Cases

### The 'Why did this change?' check
A PM sees a spike and asks the agent to find the Sales Overview dashboard, check the Revenue chart logic, and run a SQL query to see if it's a data bug or a real trend.

### The Data Audit
An engineer needs to know which databases are active. They ask the agent to list all databases and then list datasets to see what's being used in production.

### The Quick Report
An analyst needs a CSV of last month's sales. They ask the agent to find the relevant dataset and run a SQL query to pull the top 10 categories.

### Metric Discovery
A new hire wants to know how 'Churn' is calculated. They ask the agent to find the churn dashboard and explain the logic behind the main chart.

## Benefits

- Stop hunting through folders. Use list_dashboards and list_datasets to find exactly what you need in seconds.
- Understand the logic behind a metric. Use get_chart_details to see how a specific chart is calculated instead of guessing.
- Run queries faster. Use execute_sql_query to get raw data for reports without manually writing every JOIN.
- Audit your stack. Use list_databases to see every connection and ensure your data is coming from the right place.
- Skip the UI navigation. Ask your agent to find a specific dashboard and summarize its contents for you.
- Get consistent data. Ensure your agent pulls from the same validated datasets used by your official dashboards.

## How It Works

The bottom line is your agent becomes a power user of your BI platform.

1. Add the Apache Superset MCP to your client and provide your base URL and access token.
2. Connect your agent to your specific Superset instance.
3. Ask your agent to find a dashboard, explain a metric, or run a SQL query.

## Frequently Asked Questions

**Can I use the Apache Superset MCP to see my dashboards?**
Yes, it can list all of your dashboards so you can find the right one quickly without manual navigation.

**Can my agent actually run SQL queries?**
Yes, it can execute raw SQL statements via SQL Lab on any database connected to your Superset instance.

**Does this work for any BI tool?**
No, this is specifically designed for Apache Superset.

**How do I find which datasets are available?**
The agent can list all your datasets so you know exactly what's ready for analysis before you start a query.

**Can it tell me the logic behind a specific chart?**
Yes, it can pull the specific details for any chart to show you exactly how a metric is being calculated.

**Is this safe for my data?**
Yes, it uses your existing Superset credentials to perform actions just like a human user would.

**Can the AI query databases connected to Superset, such as Presto or Redshift?**
Yes. The `execute_sql_query` tool runs queries through Superset's SQL Lab API, which routes them to whichever database engine you have configured — Presto, Redshift, PostgreSQL, and others.

**Are dashboard access permissions enforced when using the Connector?**
Yes. All requests are authenticated with your `SUPERSET_ACCESS_TOKEN`, so only the dashboards, charts, and datasets your token has access to will be visible to the AI.

**Can the AI run write queries (INSERT, UPDATE, DELETE) via SQL Lab?**
It depends on your Superset database connection settings. By default, SQL Lab connections are read-only. Write access must be explicitly enabled per database in the Superset admin panel.

**Are dashboard access permissions enforced when using the MCP server?**
Yes. All requests are authenticated with your `SUPERSET_ACCESS_TOKEN`, so only the dashboards, charts, and datasets your token has access to will be visible to the AI.