What features does Bauplan MCP server provide?

Q: What features does Bauplan MCP server provide?

Data Operations: Expose commands to list tables, get schemas, retrieve specific tables, execute queries, and export results to CSV, enabling SQL-like interactions with the lakehouse through MCP.. Branch Management: Manage branches and history: list branches and commits, create and merge branches, check for existence, and delete branches.. Namespace Management: Manage namespaces within branches: list namespaces, create new ones, check existence, and delete namespaces.. Table Management: Create and manage tables from data sources, plan and apply table creation, check existence, delete tables, import data, and revert tables to a prior state.. Project Management: Run Bauplan projects from directories or code dictionaries, with configurable parameters, including dry-run, timeouts, and detach mode.. Job Management: List and inspect jobs, retrieve detailed job information, and cancel running jobs with status updates.. User Management: Query information about the currently authenticated user (e.g., username and full name).. Instructions and Guidance: Get detailed instructions for specific Bauplan use cases (pipeline, data, repair, ingest) to guide interactions and planning.

Bauplan MCP Server 2025: Features & Setup Guide

Overview

The Bauplan MCP Server is an open source library that provides AI assistants with access to Bauplan data lakehouse functionalities, including querying tables, schema inspection, data branch management, as well as running pipelines. The current release targets local development, enabling AI copilots like Claude Code or Claude Desktop to interact with your Bauplan lakehouse. The project is expected to generalize to server-side deployments in the near future, potentially hosted directly by Bauplan. It is released in beta under the MIT license, with APIs and features that may change as development continues. Quick Start guidance covers required dependencies (a Bauplan API key, uv, and an AI platform), and shows how to start the server (uv sync; uv run python main.py --transport streamable-http) and connect clients (Claude Code, Claude Desktop, Cursor). Additional tooling includes the MCP Inspector for manual testing and configuration workflows for Claude Desktop integration. Emphasizing an API-first approach, the roadmap highlights expanding deployment options, security features, and fine-grained permissions to support broader usage.

Features

Data Operations

Expose commands to list tables, get schemas, retrieve specific tables, execute queries, and export results to CSV, enabling SQL-like interactions with the lakehouse through MCP.

Branch Management

Manage branches and history: list branches and commits, create and merge branches, check for existence, and delete branches.

Namespace Management

Manage namespaces within branches: list namespaces, create new ones, check existence, and delete namespaces.

Table Management

Create and manage tables from data sources, plan and apply table creation, check existence, delete tables, import data, and revert tables to a prior state.

Project Management

Run Bauplan projects from directories or code dictionaries, with configurable parameters, including dry-run, timeouts, and detach mode.

Job Management

List and inspect jobs, retrieve detailed job information, and cancel running jobs with status updates.

User Management

Query information about the currently authenticated user (e.g., username and full name).

Instructions and Guidance

Get detailed instructions for specific Bauplan use cases (pipeline, data, repair, ingest) to guide interactions and planning.