# notbot

> notbot is an open-source AI tool for finance. Experimental Python tool scoring stock news sentiment with LLMs for public companies. It is available on CLI, API, macOS, Windows and Linux.

- **Kind:** Tool
- **Pricing:** Open source
- **Platforms:** CLI, API, macOS, Windows, Linux
- **Repository:** https://github.com/lamarcol/notbot
- **Categories:** [finance](https://ai-powered.com/categories/finance)
- **Listing updated:** 2026-11-18

## About

The README describes NotBot as a research-style Python project that gathers financial news for a chosen company or ticker and asks LLMs to rate each article's sentiment. The article scores are then combined into one weighted indicator, with weights reflecting content, source reliability and how recent each story is.

The workflow runs asynchronously: news is pulled from Google News and Finnhub, duplicate stories are filtered out, and the remaining articles are read and scored by Google Gemini or Groq models. The program finishes by writing a report that summarizes the results.

The project is aimed at developers and researchers who want to experiment with applying language models to market news. The README is explicit that its scores have not been validated against market performance and that it is not meant as investment advice. Commercial use requires separate permission from the copyright holder under a noncommercial license.

## Features

- **Multi-source news collection:** Gathers recent financial articles from Google News and Finnhub for a target company.
- **LLM sentiment scoring:** Uses Google Gemini and Groq models to rate the sentiment of each selected article.
- **Weighted aggregate score:** Combines article scores using content, source reliability and publication recency.
- **Duplicate story filtering:** Removes similar articles so repeated coverage of one event carries less weight.
- **Asynchronous processing:** Retrieves and processes multiple articles concurrently using aiohttp and asyncio.
- **Experimental price data:** Offers early support for historical end-of-day prices through Financial Modeling Prep.

## Good for

- Exploring whether news sentiment correlates with a stock's movements
- Generating a summary report of recent coverage for a given ticker
- Studying how LLMs interpret financial news in a research setting
- Prototyping sentiment pipelines in Python

## FAQ

### What does NotBot cost?

The README does not list pricing. It is released under the PolyForm Noncommercial License, and commercial use requires separate authorization from the copyright holder.

### Is NotBot suitable for investment decisions?

No. The README states it is not intended to provide financial advice or investment recommendations, and its scores have not been systematically validated against market performance.

### Which AI models does it use?

It uses Google Gemini and Groq models to evaluate news sentiment.

### How is it installed?

Clone the repository, install the requirements from PosNegNeut/requirements.txt, copy .env.example to .env, add your API keys, set the target company and ticker, then run PosNegNeut/main.py.

### Is notbot free?

Yes. notbot is open source, so you can use and self-host it for free.

### What are the best alternatives to notbot?

The closest alternatives to notbot are Notreload, Trendspider and StockGPT.

## Where this comes from

- Description: written from the project's README.

## Alternatives

- [Notreload](https://ai-powered.com/tools/notreload): Notreload is an AI tool for finance.
- [Trendspider](https://ai-powered.com/tools/trendspider): Trendspider is a paid AI tool for automation.
- [StockGPT](https://ai-powered.com/tools/stockgpt): StockGPT is a freemium AI tool for finance.

---
Source: https://ai-powered.com/tools/notbot
