Text-to-SQL Agency is a project that transforms natural language queries into SQL statements, streamlining database interactions for users. This solution is inspired by Pinterest's approach to Text-to-SQL conversion and leverages the Agency Swarm framework for orchestrating AI agents :contentReference[oaicite:0]{index=0}.
- Natural Language Processing: Converts user queries in plain English into accurate SQL commands.
- Agent-Based Architecture: Utilizes the Agency Swarm framework to manage AI agents, enhancing modularity and scalability.
- Customizable: Easily adaptable to various database schemas and user requirements.
Before setting up the application, ensure you have the following installed:
- Python 3.8 or higher
- pip (Python package installer)
- Virtualenv (optional but recommended for creating isolated Python environments)
-
Clone the Repository:
git clone https://github.com/lucklypriyansh-2/text-to-sql-agency.git cd text-to-sql-agency
python3 -m venv venv
venv\Scripts\activate
source venv/bin/activate
- #Agency Swarm Configuration:
Configure the Agency Swarm framework by setting up the necessary agents and tools as per your requirements. Refer to the Agency Swarm documentation for detailed instructions.
Start the Application:
python app/main.pyAccess the Web Interface:
- #call api
curl --location 'http://127.0.0.1:9090/completions?message=pull%20reuest%20review%20time%20vs%20no%20of%20commits&threadId=121' \
--data ''
- Sequence diagram:
.png)
.png)
Submit Queries:
Additional Resources: https://medium.com/pinterest-engineering/how-we-built-text-to-sql-at-pinterest-30bad30dabff
Enter your natural language query into the input field and submit. The application will process the input and return the corresponding SQL statement.