RuleReceipt

Hi everyone! I'm Vasiliy, a solo developer. When the original Omegle shut down, I kept thinking that the most interesting strangers are the ones you can't talk to, because you don't share a language. So I'm building Veilr: random text, voice or video chat where messages can arrive translated and calls get live captions in your language.

A few things from building it, mostly with Claude Code as a coder + critic pair of agents:

I chose captions over a dubbed voice: you still hear the real person.

An empty room kills random chat, so while you wait there's now a voice AI in your language, always labeled as AI.

When I m scrolling through a list of products, the first thing that catches my eye is usually the logo.

Only then I look at the tagline, and after that, the description.

For us, it s a very simple infinity sign, but it actually says almost everything we want our brand to say.

A loop that keeps running and never gets interrupted
Simplicity
Our brand magenta to cyan gradient

Most intros start with a bio. Mine starts with a number: 63%.

I parsed 559 public CLAUDE.md files (the rules people write for their AI coding agents). 63% of what's in them isn't an instruction at all: folder trees, notes, docs. And of the real rules, under half are concrete enough to ever check automatically.

That explained something that had been bugging me. My rules file said "never push without asking", and Claude pushed to main anyway. Nothing told me. The rules were advice, and nobody was checking.

So I built RuleReceipt. It reads your real agent sessions and tells you, rule by rule: followed, broken, or can't tell, with the exact line as proof. It only says "broken" with hard evidence. Runs locally; nothing leaves your machine.