Meet Ragstar

Your AI Data Analyst for dbt Projects

Beta Python 3.10+ MIT License Black MyPy Ruff
View on GitHub Get Started

Key Features

Powerful AI capabilities for your dbt project

Natural Language Q&A

Ask questions about models, sources, metrics, lineage, and more in plain English. Get answers that make sense, not just data dumps.

Agentic Intelligence

Ragstar intelligently analyzes dbt models, understanding logic and context to provide insights that matter to your team.

Automated Documentation

Automatically generate model and column descriptions where they're missing, improving data catalog quality.

Semantic Search

Find relevant assets based on meaning, not just keywords, making data discovery intuitive and powerful.

Slack Integration

Built-in Slackbot for easy interaction, bringing data insights directly to your team's workflow.

Feedback Loop

Track questions and feedback to continuously improve answers and recommendations over time.

Use Cases

How teams leverage Ragstar to improve data workflows

Accelerate Data Discovery

Quickly find relevant dbt models and understand their purpose without digging through code.

Improve Onboarding

Help new team members understand the dbt project structure and logic faster.

Maintain Documentation

Keep dbt documentation up-to-date with automated generation and suggestions.

Enhance Data Governance

Gain better visibility into data lineage and model dependencies.

Debug dbt Models

Ask clarifying questions about model logic and calculations.

Architecture

How Ragstar works under the hood

Key Components:

  • dbt Project Parsing: Extracts comprehensive metadata from dbt artifacts or source files.
  • PostgreSQL + pgvector: Central knowledge store with vector embeddings for semantic search.
  • Vector Embeddings: Numerical representations of model documentation capturing semantic meaning.
  • Large Language Models: Powers natural language understanding and generation.
  • Agentic Reasoning: Step-by-step reasoning process for model interpretation.
  • CLI Interface: Command-line tools for interacting with the system.

Setup & Installation

Get started with Ragstar in minutes

Docker Compose (Recommended)

The easiest way to get started with Ragstar is using Docker Compose.

  1. Clone the repository: 
    git clone https://github.com/pragunbhutani/ragstar.git
cd ragstar
  2. Configure environment variables: 
    cp .env.example .env
# Edit .env with your API keys and configuration
  3. Build and run with Docker Compose: 
    docker compose up --build -d
  4. Initialize with dbt project data: 
    docker compose exec app ragstar init cloud
# or
docker compose exec app ragstar init local
# or
docker compose exec app ragstar init source
  5. Build the knowledge base: 
    docker compose exec app ragstar embed --select "*"

Local Python Environment

For advanced users who prefer not to use Docker.

  1. Check prerequisites:
    • Python 3.10+
    • Poetry
    • PostgreSQL with pgvector
  2. Clone and install: 
    git clone https://github.com/pragunbhutani/ragstar.git
cd ragstar
poetry install
  3. Configure environment variables: 
    cp .env.example .env
# Edit .env with your configuration
  4. Initialize the database: 
    poetry run ragstar init-db
  5. Load dbt project data: 
    poetry run ragstar init cloud
# or
poetry run ragstar init local

Ready to democratize your data?

Ragstar makes it easy for anyone to ask questions about your dbt project and get intelligent answers.

View on GitHub Get Support