---
license: mit
sdk: streamlit
sdk_version: 1.55.0
---
# 🧠 DataMind Agent
### AI-Powered Data Analyst — LangChain + Gemini + Streamlit

Upload any data file (CSV, Excel, JSON) and chat with your data using natural language. The agent analyzes, visualizes, and explains your data powered by Google Gemini.

---

## 🚀 Features

| Feature | Description |
|---|---|
| 📂 Multi-format support | CSV, Excel (.xlsx/.xls), JSON |
| 💬 Natural language Q&A | Ask anything, get intelligent answers |
| 📊 Auto visualizations | AI picks the best chart for your question |
| 🎨 Custom chart builder | Build any chart with dropdown controls |
| 🔍 Data explorer | Filter, search, and download raw data |
| 🧠 AI data summary | Executive summary generated by Gemini |

---

## 📁 Project Structure

```
data-analyst-agent/
├── app.py              # Streamlit UI (main app)
├── core_agent.py       # LangChain + Gemini logic
├── requirements.txt    # Python dependencies
├── .env                # API key config
├── sample_data.csv     # Test dataset (sales data)
└── README.md           # This file
```

---

## ⚙️ Setup & Installation

### Step 1 — Clone / download the project
```bash
cd data-analyst-agent
```

### Step 2 — Create a virtual environment (recommended)
```bash
python -m venv venv

# On Windows:
venv\Scripts\activate

# On Mac/Linux:
source venv/bin/activate
```

### Step 3 — Install dependencies
```bash
pip install -r requirements.txt
```

### Step 4 — Get your free Gemini API key
1. Go to [https://aistudio.google.com/app/apikey](https://aistudio.google.com/app/apikey)
2. Sign in with Google
3. Click **"Create API Key"**
4. Copy the key (starts with `AIza...`)

### Step 5 — Add your API key
Either paste it directly in the app sidebar, OR add it to `.env`:
```
GOOGLE_API_KEY=AIzaYourKeyHere
```

### Step 6 — Run the app
```bash
streamlit run app.py
```

The app opens at **http://localhost:8501**

---

## 🎯 How to Use

1. **Paste your Gemini API key** in the sidebar
2. **Upload a data file** (CSV, Excel, or JSON)
3. **Dashboard tab** — see auto-generated stats and charts
4. **Chat tab** — ask questions like:
   - *"What are the top selling products?"*
   - *"Is there a correlation between age and spending?"*
   - *"Show me outliers in the sales column"*
5. **Charts tab** — build custom visualizations
6. **Raw Data tab** — filter and download your data

---

## 💡 Example Questions to Ask

```
"What is the average profit by category?"
"Which region has the highest sales?"
"Are there any missing values I should worry about?"
"What trends do you see in the data over time?"
"Which customers are the most valuable?"
"Give me a statistical summary of all numeric columns"
"What correlations exist between the columns?"
```

---

## 🏗️ Architecture

```
User (Streamlit UI)
       │
       ▼
  app.py (UI Layer)
       │
       ├── core_agent.py
       │       ├── load_file()          → Parses CSV/Excel/JSON → DataFrame
       │       ├── profile_dataframe()  → Statistical profiling
       │       ├── ask_agent()          → LangChain → Gemini → Answer
       │       ├── make_plotly_chart()  → Renders visualizations
       │       └── ai_recommend_chart() → Gemini picks best chart
       │
       └── Google Gemini 1.5 Flash (via LangChain)
```

---

## 📦 Key Libraries Used

| Library | Purpose |
|---|---|
| `langchain` | Agent framework, prompt management |
| `langchain-google-genai` | Gemini LLM integration |
| `streamlit` | Web UI |
| `pandas` | Data loading and manipulation |
| `plotly` | Interactive visualizations |
| `openpyxl` / `xlrd` | Excel file support |

---

## 🔧 Customization Ideas

- Add **PDF support** using `pdfplumber`
- Add **database connection** (SQLite, PostgreSQL)
- Add **export to PowerPoint** for chart reports
- Add **multi-file comparison** mode
- Deploy to **Streamlit Cloud** (free hosting)

---

## 🆓 Free Tier Limits (Gemini 1.5 Flash)
- 15 requests per minute
- 1 million tokens per minute
- 1,500 requests per day

This is more than enough for personal data analysis projects!

---

*Built with ❤️ using LangChain + Google Gemini + Streamlit*