> ## Documentation Index
> Fetch the complete documentation index at: https://sourcetable.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# API Playground

> Test, configure, and connect APIs and databases through the Superagents interface.

The API Playground lets you test connections before using them with the AI. You can verify authentication, preview responses, and configure which resources are accessible.

## Adding a database connection

<Steps>
  <Step title="Open Superagents">
    Click the **Superagents** button in the toolbar or use the connectors menu in the AI chat.
  </Step>

  <Step title="Select Database">
    Choose your database type: PostgreSQL, MySQL, MongoDB, Snowflake, SQL Server, or any of the 25+ supported databases.
  </Step>

  <Step title="Enter credentials">
    Provide host, port, database name, username, and password. Or paste a connection string.
  </Step>

  <Step title="Test connection">
    Click **Test** to verify the connection works. Sourcetable validates connectivity and lists available tables.
  </Step>

  <Step title="Save">
    Save the connection. It's now available to the AI for natural language queries.
  </Step>
</Steps>

## Adding an API connection

<Steps>
  <Step title="Select API">
    Choose from the pre-built connector list or add a custom API.
  </Step>

  <Step title="Authenticate">
    Enter your API key, bearer token, or configure OAuth. Different APIs support different auth methods.
  </Step>

  <Step title="Configure endpoints">
    Select which endpoints the AI can access. You can restrict to read-only or allow write operations.
  </Step>

  <Step title="Test">
    Make a test request to verify the API responds correctly.
  </Step>
</Steps>

## Adding an MCP connection

MCP (Model Context Protocol) connections let you add specialized tools:

<Steps>
  <Step title="Select MCP">
    Choose from available MCP servers (e.g., Apify for web scraping).
  </Step>

  <Step title="Configure">
    Enter the MCP server URL and any required authentication.
  </Step>

  <Step title="Select tools">
    Choose which tools from the MCP server the AI can use.
  </Step>
</Steps>

## Supported databases

| Category   | Databases                                                           |
| ---------- | ------------------------------------------------------------------- |
| PostgreSQL | Generic, AWS Aurora, AWS RDS, Azure, Google Cloud, Heroku, Supabase |
| MySQL      | Generic, AWS Aurora, AWS RDS, Azure, Google Cloud                   |
| MariaDB    | Generic, AWS RDS, Azure                                             |
| MongoDB    | Standard, Sharded Cluster                                           |
| SQL Server | Generic, Azure, AWS RDS                                             |
| Other      | Snowflake, DynamoDB, Cosmos DB, Magento                             |

## Using connections with the AI

Once connected, ask the AI questions about your data:

```
"Show me the top 10 customers by revenue from the production database"
```

```
"Pull the latest campaign metrics from the Google Ads API"
```

```
"Scrape the pricing table from competitor.com using Apify"
```

The AI automatically selects the right connection and formats the query.