Anthropic-SQL-Agent(A-SQL-A)

👋 Introduction 👋

Hello! I am a beginner developer who is greatly interested in the rapidly emerging field of LLMs.

This is my another basic starter project, created for the purpose of studying LLM prompting and Python for my own study.

Thank you for coming by, and please keep an eye out for future updates!

🌠 OverView 🌠

This project is a simple starter project that demonstrates how you can utilize LangChain and Anthropic's Claude model to perform Q&A on your personal SQL Database.

🚀 HOW TO START 🚀

1. Clone this repository.

git clone https://github.com/MIRACLE-cowf/A-SQL-A.git

2. Move to the cloned repository.

cd A-SQL-A

3. Inside the A-SQL-A folder, fill in the necessary API keys in the .env file.
4. Place the CSV file that you want to convert into a database inside the src/csv folder within the A-SQL-A folder.
5. Install the required libraries

pip install -r requirements.txt

6. Run main.py

py -m main

• Note: If you put a CSV file inside the src/csv folder and run main.py, it will automatically convert the CSV file into a .db file.

✅ Check Out My Other Project ✅

🔥 PAR(Powerful-Agent-Researcher) 🔥

This project is an Agent Assistant that aims to automatically generate high-quality documents in response to user question. It conducts its own web searches and creates documents to provide answers.

Main Features

• THLO(Think-High-Level-Outline) : It deeply 'think' and 'inner monologue' to understands user question and automatically generates high-level-outline or plan for document creation.
• Utilization of Various Information Sources : It integrates multiple search engines, including Tavily, YouTube, arXiv, and Wikipedia, to gather comprehensive and intelligent information.
• Intelligent Information Summarization and Integration : It intelligently summarizes and integrates the collected information to automatically generate documents that are not only optimized for the user's question but also contain richer and high-quality information.

If you're interested, check it out HERE
I'm actively seeking feedback and discussions!