Launching today

Claude Code OS

Launching today

52 commands, 12 agents & 9 safety hooks for Claude Code

4 followers

52 commands, 12 agents & 9 safety hooks for Claude Code

4 followers

Visit website

Claude Code OS makes Claude Code behave like a disciplined senior team: it plans before coding, writes tests first, reviews its own diffs — and physically blocks secrets, `git commit --no-verify`, and force-pushes to main at the tool level. Free + MIT core; optional 52-command PRO.

Free Options

Launch tags:Open Source•Developer Tools•Artificial Intelligence

Launch Team

Checkmarx Next Generation SASTHighest Fidelity F1 Score Hybrid Engine, Language Agnostic

Promoted

Maker

📌

Hey Product Hunt 👋 I built this because Claude Code, out of the box, codes like a brilliant junior who never read your repo: it implements before planning, skips tests unless you ask, and will happily `git commit --no-verify` straight past a failing pre-commit hook or paste an API key into a source file. The model isn't the problem — the setup is. So I built the missing operating layer: - /plan won't write a line until you type CONFIRM - /verify runs build → types → lint → tests → secret-scan into one PASS/FAIL scoreboard - 8 safety hooks block secrets, interpolated SQL, force-pushes, and --no-verify at the tool level — not "please behave," an actual wall (~30 lines of zero-dependency Node each, 61 tests) - 12 agents with model routing baked in (Opus where it pays, Haiku where it's cheap) The free repo is MIT and installs in 60 seconds (backs up your config, never overwrites your files). There's a paid 52-command superset, but the free tier stands on its own. I'd genuinely love feedback on two design calls: should a security hook fail open or closed, and should the --no-verify block have any escape hatch at all? Happy to answer anything.

Report

7d ago

Forum Threads

p/claude-code-os

•

3d ago

What is the scariest thing your AI coding agent has ever done?

I have been bolting guardrails onto my Claude Code setup for a few months now, mostly out of self defense. The moment that did it for me was watching it run git commit --no-verify straight past a failing pre-commit check, like the check was just a friendly suggestion it could ignore. Then a week ago it got stuck retrying the same broken build command in a loop while I was away from my desk, quietly eating tokens the whole time.

So before I launch my own take on this next Tuesday, I would rather hear yours first. What is the scariest or most expensive thing your AI coding agent has actually done to a real project? And if you could give it one hard rule it could never break, what would it be?

View all