LogSpotAI

A lightweight AI-assisted log analysis tool written in Python. LogSpotAI helps you detect suspicious or anomalous log entries using a large language model (LLM) or other AI methods. The entire project is kept minimal, with just one or two Python files, so it's easy to understand, modify, and extend.

Features

• Simple CLI: Pass in your log file, get back suspicious lines.
• AI-Powered: Uses OpenAI's API (or switch to your preferred AI model).
• Minimal: Only one main Python script + this README + requirements.
• Extendable: You can adapt it for various log formats and different AI endpoints.

Getting Started

1. Clone the Repository

git clone https://github.com/heinrihs-s/LogSpotAI.git
cd LogSpotAI

2. Create a Virtual Environment

# On Linux/Mac
python -m venv venv
source venv/bin/activate

# On Windows
python -m venv venv
.\venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Set Up Your OpenAI API Key

• Sign up or log in to OpenAI to get an API key
• Export the key as an environment variable:

# Linux/Mac
export OPENAI_API_KEY="your_secret_key"

# Windows (Command Prompt)
set OPENAI_API_KEY="your_secret_key"

Or create a .env file (if you add python-dotenv support) to store your key locally.

5. Usage

Basic usage:

python ailogger.py --logfile /path/to/your.log

Example with sample log:

python ailogger.py --logfile sample.log

• The script reads your logs, batches them, and queries the AI model to identify potentially suspicious lines
• Suspicious lines are then printed out for review

Customization

• Local Models: If you prefer not to use the OpenAI API, switch to a local anomaly-detection model (e.g., scikit-learn's IsolationForest or a small transformer)
• Thresholding: For local models, you can tweak detection thresholds to adjust sensitivity
• Log Parsing: Enhance the parse_logfile function to handle different log formats (JSON logs, Apache logs, etc.)
• Report Generation: Output results to a separate file, or incorporate visualizations if you want

Contributing

Contributions, issues, and feature requests are welcome! Feel free to open a PR or issue on the GitHub repository.

License

This project is licensed under the MIT License. You are free to use, modify, and distribute this software as you see fit, but attribution is appreciated.

---

Additional Tips

1. If you want to keep everything in just one Python file, you can place your CLI and AI logic all in ailogger.py and skip any separate scripts
2. Add a sample log file (sample.log) and demonstrate how the script flags suspicious lines
3. Consider adding a small test script or a test folder with example inputs and outputs

Happy Logging with LogSpotAI!