Launched this week

CodeCrab

Launched this week

100% local-first AI pull request reviewer

54 followers

100% local-first AI pull request reviewer

54 followers

Visit website

Code Review Tools

CodeCrab is a native desktop app that reviews Pull Requests in seconds using your local CLIs. 100% client-side — your code never leaves your machine.

Free

Launch tags:Software Engineering•Developer Tools•Artificial Intelligence

Launch Team / Built With

CodeCrab

Maker

📌

I built CodeCrab as a 100% Local-First desktop app to fix my own code review routine. Now with AI we generate code super fast, but reviewing the mountain of Pull Requests that get created became a bottleneck. To solve my own workflow I built CodeCrab, a desktop app to speed up code reviews with AI 100% locally. I use CodeCrab every day and it greatly increased my code review speed. 🔒 100% Local-First: Your code never leaves your machine and is never uploaded to third-party servers. 💰 No extra costs: Directly reuse your Claude Code tokens or subscription. ⚡ Parallel reviews: Run 5, 10, or more reviews at the same time in the background while you do something else. 🎯 Human-in-the-loop: Shows you bugs and missing tests in a clean interface so you make the final decision.

Report

4d ago

Opengeni

Really cool! Does it also use local LLM models? What do you recommend hardware wise for that to work 32gb macbook m5 etc or better specs? Whats your experience around this?

Report

3d ago

CodeCrab

Maker

@alchemist_studio Yes, CodeCrab supports local LLMs! You reuse your subscriptions/API keys from Claude Code for now, with Cursor, Codex, and others coming soon. Because it's built with Tauri, it’s super lightweight, fast, and uses very little memory—so it runs smoothly on pretty much any hardware setup!

Report

3d ago

Forum Threads

p/codecrab

•

2h ago

Is Code Review becoming a total bottleneck in your team, or is AI making it worse?

As teams ship faster with AI coding assistants (Claude Code, Cursor, etc.), the volume of PRs has skyrocketed. But the time engineers spend reviewing code hasn't scaled at all.

I've seen two sides to this lately:

Teams where PRs sit in queue for 2 3 days, creating a massive pipeline bottleneck.
Teams trying AI reviewers, but ending up with automated spam comments that nobody reads.

I m curious how your team handles this:

View all