Tolerance uses an LLM to score social media posts by manipulation tactics (rage bait, engagement farming, emotional exploitation) and progressively blurs high-scoring content. The more a post tries to grab your attention, the harder it is to see.

I built it because I'd been fighting a doomscrolling habit for 15 years. Blockers never worked, I'd just override them. Tolerance broke the habit in two weeks. I still reflexively open Twitter and Reddit, but I close them in seconds because there's nothing rewarding there anymore.

I started feeling entitled to my own attention. Manipulation that I used to scroll past now feels like a violation. The less a post tries to grab you, the higher the actual value tends to be.

Open source (AGPL-3.0), free tier with BYOK, $9/month pro tier.

https://tolerance.lol https://github.com/gtech/tolerance