Tickermood

Tickermood is a Python package that provides market sentiment analysis for stock tickers based on news from multiple sources. It combines web scraping techniques with large language models (LLMs) using the LangChain and LangGraph frameworks to generate sentiment scores for given tickers.

📦 Installation

Install Tickermood via pip:

pip install tickermood

Note: To use Tickermood locally, Ollama must be installed and qwen3:4b model needs to be available.

ollama pull qwen3:4b

🚀 Usage

Programmatic Usage

from tickermood import TickerMood

ticker_mood = TickerMood.from_symbols(["AAPL", "GOOGL", "MSFT"])
ticker_mood.run()

CLI Usage

tickermood AAPL GOOGL MSFT

This will:

• Fetch the latest news for the specified tickers
• Run LLM agents to analyze the news
• Provide a sentiment score for each ticker

Results are stored in a SQLite database.

🗃️ Database

Tickermood creates a SQLite database in the current directory named tickermood.db if it doesn't already exist. It includes:

• Sentiment ratings (e.g., Buy, Hold, Sell)
• Price targets
• Summaries of the fetched news articles

⚙️ LLM Backend Options

Default: Local LLM (Ollama)

• Runs LLMs locally for free
• Performance depends on your hardware

Optional: OpenAI API

• Requires setting the OPENAI_API_KEY environment variable

Or, pass the key via CLI:

tickermood AAPL GOOGL MSFT --openai_api_key_path /path/to/openai_api_key.txt

📝 License

MIT License