📈 Stockly.AI

A sharp, no-fluff AI stock market assistant — built with Streamlit, powered by Ollama cloud models and real-time Yahoo Finance data.

---

Overview

Stockly.AI is a conversational stock analysis web app. Users can ask natural-language questions about any stock and get concise, data-backed answers alongside a live candlestick chart, key financial metrics, and the latest news headlines — all within a sleek dark-mode interface.

---

Features

• AI Chat Interface — Conversational Q&A powered by Ollama cloud models (streaming responses)
• Real-Time Stock Data — Live prices, market cap, P/E, EPS, 52-week high/low, beta, dividend yield, and analyst ratings via yfinance
• Candlestick Charts — Interactive OHLC chart with volume overlay via Plotly, supporting 1W / 1M / 3M / 6M / 1Y / 5Y timeframes
• Auto Ticker Detection — Automatically recognizes stock tickers mentioned in chat (e.g. "What do you think about NVDA?") from a curated list of 100+ popular symbols
• Latest News — Fetches and displays the 4 most recent headlines for the active ticker
• Quick Lookup Sidebar — Type any ticker, pick a time range, and pull up the chart instantly
• Popular Stocks Shortcuts — One-click buttons for AAPL, TSLA, NVDA, MSFT, AMZN, META, GOOGL, JPM
• Dark Aesthetic UI — Custom CSS styled after Claude.ai's dark theme using Inter font

---

Tech Stack

Layer	Technology
Frontend / App	Streamlit
AI Backend	Ollama Cloud API (streaming)
Stock Data	yfinance
Charts	Plotly
HTTP	requests
Data	pandas

---

Project Structure

.
├── app.py            # Main Streamlit application
├── requirements.txt  # Python dependencies
└── README.md

---

Getting Started

Prerequisites

• Python 3.9+
• An Ollama account with an API key

Installation

# 1. Clone the repo
git clone https://github.com/your-username/stockly-ai.git
cd stockly-ai

# 2. Install dependencies
pip install -r requirements.txt

# 3. Add your Ollama API key to Streamlit secrets
mkdir -p .streamlit
echo 'OLLAMA_API_KEY = "your_key_here"' > .streamlit/secrets.toml

# 4. Run the app
streamlit run app.py

Then open http://localhost:8501 in your browser.

---

Configuration

Streamlit Secrets

The app reads your Ollama API key from Streamlit's secrets manager. Create .streamlit/secrets.toml:

OLLAMA_API_KEY = "your_ollama_api_key_here"

When deploying to Streamlit Community Cloud, add this key under App Settings → Secrets.

AI Model

The active model is selected from the sidebar at runtime. To add or change models, edit the CLOUD_MODELS dictionary in app.py:

CLOUD_MODELS = {
    "GPT-OSS 120B": "gpt-oss:120b",
    # Add more Ollama-compatible model IDs here
}

AI Behavior

Two constants control AI response style:

TEMPERATURE = 0.5   # Creativity level (0 = deterministic, 1 = creative)
MAX_TOKENS  = 1000  # Max response length

The system prompt (SYSTEM_PROMPT) is tuned for concise, data-first stock analysis. Modify it in app.py to adjust the assistant's tone or behavior.

---

Usage

1. Chat — Type any question in the chat bar, e.g.:

   • "Analyze Apple stock for me"
   • "Will NVDA go up this month?"
   • "Compare AAPL vs MSFT vs GOOGL"
   • "What's the latest news on Tesla?"

2. Quick Lookup — Enter a ticker in the sidebar and click Show Chart to pull up metrics and the candlestick chart.

3. Popular Stocks — Click any shortcut button in the sidebar (AAPL, TSLA, NVDA, etc.) for instant chart access.

4. Clear / Reset — Use the sidebar buttons to clear chat history or reset to the default AAPL view.

---

Deployment

Streamlit Community Cloud (recommended)

1. Push the repo to GitHub.
2. Go to share.streamlit.io and connect your repo.
3. Set OLLAMA_API_KEY under App Settings → Secrets.
4. Deploy.

Docker

FROM python:3.11-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
EXPOSE 8501
CMD ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]

---

Caching

Stock data is cached with a 5-hour TTL (ttl=18000) using @st.cache_data to avoid redundant API calls to Yahoo Finance. To force a refresh, clear the Streamlit cache via the app menu or restart the server.

---

Disclaimer

Stockly.AI is for informational and educational purposes only. Nothing in this application constitutes financial advice. Always do your own research before making investment decisions.

---

License

See LICENSE for details.