Balon - LLM-assisted content blocker

Balon filters social media posts through an LLM to help you block content you don't want to see.

Here's how it works :

- Craft a prompt that explains the things you rather not see. For example :

If the following text includes any negative opinion towards paellas, return "paella".

- Make sure that your prompt asks for one word (to save tokens & time) and for ok if other conditions are not met. So your prompt could be :

Answer only with one word. If the following text includes any negative opinion towards paellas, return "paella". If it is about pomegranates or other sour fruits, return "pomegranate". Otherwise return "ok".