Databot / README.md
julkarnaeen's picture
Upload README.md with huggingface_hub
e67b4b5 verified
---
title: DataBot ERP Assistant
emoji: πŸ€–
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: "5.20.0"
app_file: app.py
pinned: false
license: mit
---
# DataBot – ERP Database Assistant
DataBot lets you ask **natural-language questions** about your ERP database.
It uses GPT-4o to translate your question into SQL, executes it safely on your
MySQL database, and returns a plain-English summary.
## Features
- Natural-language β†’ SQL β†’ plain-English answers
- Whitelist-based table access control (`data_access_config.json`)
- Blocks all write operations (INSERT / UPDATE / DELETE / DROP …)
- Masks sensitive columns (passwords, IBAN, salary, PII)
- Schema cached at startup for fast responses
## Setup (Hugging Face Spaces)
Set the following **Secrets** in your Space settings
(*Settings β†’ Variables and secrets β†’ New secret*):
| Secret name | Description |
|----------------|---------------------------------------|
| `DB_USER` | MySQL username |
| `DB_PASSWORD` | MySQL password |
| `DB_HOST` | MySQL host IP / domain |
| `DB_NAME` | Database name (default: `dev_poly`) |
| `OPENAI_API_KEY` | Your OpenAI API key |
| `LLM_MODEL` | Model to use (default: `gpt-4o`) |
## Local Development
```bash
# 1. Clone the repo and create .env
cp .env.example .env # fill in your credentials
# 2. Install dependencies
pip install -r requirements.txt
# 3. Run locally
python app.py
```
Then open http://localhost:7860 in your browser.