# BEEM Data Platform

> BEEM is a data platform that lets you connect data sources, build SQL datasets, create visualizations and dashboards, and manage data workflows — all through a web interface or programmatically via an MCP server.

## MCP Server

BEEM provides a Model Context Protocol (MCP) server that allows AI assistants (Claude Desktop, Claude Code, claude.ai, or any MCP-compatible client) to interact with the platform programmatically.

Add the remote MCP server URL to your MCP client (Claude Desktop, Claude Code, claude.ai, or any MCP-compatible client):

- https://mcp.beemdata.com/mcp

Authentication is handled via the standard OAuth2 flow built into the MCP protocol.

For full setup instructions, see: https://docs.beemdata.com/docs/setup-guide

## Available MCP Tools

The MCP server exposes the following capabilities:

### Workspaces & Organizations
- List and get workspaces
- List user organizations
- Get current user info

### Folders
- Create, list, get, and update folders

### Datasets
- Create, list, get, rename, and update datasets
- Save and deploy SQL queries (two-phase: save draft, then deploy)
- Preview SQL query results
- Query dataset content (rows)
- Refresh dataset data

### Dataset Tests
- Create, list, get, update, and delete dataset tests
- Save and deploy test SQL queries

### Sources & Data Objects
- List and get sources (data catalogs/connections)
- List and get data objects (tables, views) within sources
- Create data objects and start ingestion

### Dashboards
- Create, list, get, and update dashboards
- Create, list, get, and update dashboard widgets

### Visualizations
- Create, list, get, update, and delete visualizations

### Flows
- List and get data flows (pipelines/workflows)

### Events & Monitoring
- List events by workspace, user, or object
- Get DPU (Data Processing Unit) usage

## Key Concepts

- **Workspace**: A project-level container for datasets, dashboards, and sources.
- **Dataset**: A SQL-defined data model. Datasets use a two-phase deployment: save the SQL query as a draft, then deploy it (which creates the underlying Redshift view and scheduling).
- **Source**: A data connection/catalog (e.g., a database, S3 bucket, or API).
- **Data Object**: A specific table or view within a source.
- **Folder**: Organizes datasets and other assets within a workspace.
- **Dashboard**: A collection of widgets displaying visualizations.
- **Visualization**: A chart or table built on top of a dataset.
- **Flow**: A data pipeline that orchestrates ingestion and transformation steps.
- **DPU**: Data Processing Unit — the unit of compute consumption.

## App URL

- https://app.beemdata.com