kevindutra

kevindutra / crit

Public

TUI for reviewing AI-generated code and plans

38
3
100% credibility
Found Mar 14, 2026 at 38 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

A terminal user interface for humans to review and annotate AI-generated code changes or plans, saving comments in a format consumable by AI tools.

How It Works

1
🔍 Discover the review helper

You hear about a handy tool that lets you easily check AI-written plans or code changes right in your terminal.

2
📥 Get it ready

You quickly add it to your setup, either through your AI chat tool or directly on your computer.

3
🚀 Start a review session

You pick a file like an AI plan or your recent code updates, and a nice screen opens showing everything clearly.

4
Choose what to review
📝
Review a plan

Dive into one document, scroll through, and note feedback where it matters.

🔄
Review code changes

Switch between files, see what's new or different, and comment on the updates.

5
💬 Add your thoughts

Use simple keys to jump around, select lines, and drop inline notes exactly where you spot issues—it feels smooth and intuitive.

6
💾 Finish and share

Close the screen, and your comments are saved ready for the AI to read and act on.

✅ AI improves perfectly

Your AI buddy uses your feedback to tweak the code or plan, making it just right on the next try.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 38 to 38 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is crit?

Crit is a Go-powered TUI for reviewing AI-generated code diffs and plans right in your terminal. Fire up `crit review --code` to tab through git changes with syntax highlighting, diff markers, and inline comments; or `crit review plan.md` for single-file markdown scrutiny. It saves feedback as JSON via `crit status`, feeding back to agents like Claude for seamless human-in-the-loop iteration.

Why is it gaining traction?

In the awesome TUI GitHub ecosystem of github tui apps and github tui terminals, crit shines with tmux split-pane mode for side-by-side Claude Code reviews, snappy keybindings (j/k navigate, v for visual select, [/] jump comments), and scriptable CLI like `crit comment file.md --line 10 --body "fix this"`. No fluff – just fast feedback on ai-generated outputs without alt-tabbing.

Who should use this?

Backend devs iterating Claude-generated refactors across git files, AI prompt engineers tweaking markdown plans, or terminal diehards auditing agent code before commit. Ideal for anyone in agentic workflows needing quick, structured critiques.

Verdict

Worth `go install` for AI coding loops – polished CLI, tmux integration, and docs punch above 38 stars and 1.0% credibility score. Early but battle-tested with good coverage; it'll hit critical mass fast if Claude users pile in.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.