LNChat is a reckless experiment that combines the power of OpenAI's GPT-4 model with the capabilities of managing a Lightning node.
This application directly executes the Python code generated by the GPT-4 model in response to your queries. While we've tried to ensure safety, there are no guarantees. There is a real risk that errors or malicious responses from the AI could cause you to lose funds.
Use LNChat at your own risk. If you're not comfortable with this level of risk, do not use this application.
- Python 3.6+
- Flask
- An OpenAI API key
- A Lightning node
- Clone the repository.
- Install the dependencies with
pip install -r requirements.txt. - Get ahold of your node's macaroon and cert (maybe from nodes.voltage.cloud/) and put them somewhere handy.
- Copy the
.env.sampleto.envand fill in your environment variables. - Run the application with
python app.py.
- Navigate to
http://127.0.0.1:5555in your web browser. - Enter a query about managing your Lightning node. For example, "How many peers are connected to my node?"
- Submit the query and view the response.
Remember, this application is #reckless. It's fun to experiment with, but use it responsibly and at your own risk.