Every time I try to learn something useful, my brain decides to launch a completely unrelated side quest. I’m not sure whether this is generational brain rot, my ADHD leveling up, or me simply being incredibly lazy.
The plan: review calculus for one hour. The outcome: spend three and a half hours learning how nuclear submarines make oxygen.
So, naturally, instead of developing discipline like a functional adult, I procrastinated by building an entire app to manage the problem for me.
It’s a customizable forward proxy where you create rules as nodes, connect them visually into policy flows, and group those flows into different modes: https://github.com/Vaccarini-Lorenzo/ProductivityProxy
I originally made it for myself, but work and other boring adult responsibilities have eaten most of my free time. Rather than letting the project quietly rot on my computer, I’m putting it out there in the hope that someone smarter—or simply less busy—might turn it into something genuinely good. Or maybe it’s already useful to someone. That would also be nice.
A couple of things worth knowing:
- It’s aimed at “power users.” Some nodes may require a bit of Python to define their behavior. Fortunately, we now live in a world where you can politely ask an LLM to write it for you. The project is documented, and I’m planning to add an AGENTS.md file and agent-friendly APIs. Eventually, you should be able to describe the flow you want in plain language and let your favorite LLM handle the policies.
- It runs on top of mitmproxy, so you’ll need that installed, including its certificates.
P.S. The UI is built with Tauri. Automatic proxy configuration currently works only on macOS, and I’ve only tested it on Apple Silicon. On other platforms, you can still use it, but you’ll need to configure your browser’s proxy settings manually. Proper cross-platform support remains firmly in TODO territory.