QueryCraftAI is an open-source, multi-agent AI system that turns plain English questions into optimized SQL queries — and explains the logic behind them.
It bridges the gap between complex relational schemas and human-friendly natural language, enabling developers, analysts, and business users to explore data conversationally without writing a single line of SQL.
Vision: Democratize data access — make querying as simple as having a conversation.
- Multi-Agent Orchestration: Modular agentic pipeline — intent detection, table identification, column pruning, SQL generation, and explanation — each powered by structured LLM calls.
- Context-Aware Querying: Maintains chat history and dynamically enhances prompts to handle follow-ups and modifications intelligently.
- Schema-Constrained Generation: Every SQL is schema-validated — no hallucinated tables or columns.
- Explainable AI Queries: Each generated SQL comes with a concise, human-readable explanation.
- Open Source & Extensible: Easily integrate with any SQL database, enhance with RAG, or plug in your own agents.
The application uses a multi-agent system where each agent has a specialized role. The process is orchestrated by a backend Flask API and presented to the user through a simple frontend.
- Rapid Prototyping: Developers can quickly generate complex queries needed for new application features.
- Data Exploration: Analysts can perform ad-hoc analysis without writing boilerplate SQL.
- Business Intelligence: Business users can get answers to questions like "How many new users signed up last week?" without waiting for an analyst.
- Learning SQL: Junior developers or students can use the agent as a tool to learn SQL by seeing how their questions translate into code.
To get a local instance of the agent running, follow these steps:
- Install dependencies:
pip install -r requirements.txt
- Run the development server:
./devserver.sh
- Open your browser and navigate to the local URL provided.
This project is just the beginning. Our future plans include:
- Support for More SQL Dialects: Adding support for PostgreSQL, MySQL, and others.
- Data Visualization: Automatically generating charts and graphs to visualize the query results.
- Query History & Saving: Allowing users to save and reuse frequently asked questions.
- Advanced Data Context: Enabling the agent to understand more complex relationships and business-specific logic.
- Integration with BI Tools: Connecting the agent to popular BI platforms.
This is an open-source project, and we welcome contributions from the community! Whether it's a bug fix, a new feature, or improved documentation, your help is valued.