tomasz-tomczyk

tomasz-tomczyk / crit

Public

GitHub PR-style review tool for your LLM agent's markdown output

crit.live agentic-coding ai-tools cli llm markdown
23
2
100% credibility
Found Feb 20, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Crit is a lightweight desktop app that opens a browser interface for adding GitHub PR-style inline comments to Markdown plans, generating an interleaved review file for iterative feedback with AI agents.

How It Works

1
💡 Discover Crit

You find a handy tool to review plans written by AI assistants, adding notes just like checking changes on a shared document.

2
📥 Set up Crit

Download the ready-to-use app and put it in a spot where you can launch it anytime.

3
📄 Open your plan

Point the tool at your Markdown plan file, and it instantly opens a clear, easy-to-read view in your web browser.

4
✏️ Leave inline notes

Click any line or section to add your feedback, questions, or ideas right next to the text, feeling like a collaborative edit session.

5
🔄 Build the review

Watch as your notes weave into the plan in real time, creating a complete reviewed version you can hand back.

6
✅ Finish and loop

Hit done to grab the full reviewed plan and a ready message for your AI; it auto-refreshes when the plan updates for the next round.

🚀 Smarter plans emerge

Your AI absorbs the feedback, refines the plan round after round, and you end up with polished, perfect results effortlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 23 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 CLI tool that turns markdown plans from LLM agents into interactive GitHub PR-style reviews. Run `crit plan.md` to spin up a local web UI where you add inline comments on single lines or blocks, with real-time updates to a `.review.md` file interleaved with the original content. It handles syntax highlighting, Mermaid diagrams, dark mode, and auto-reloads when the source file changes—perfect for multi-round iteration with AI agents like those in GitHub Copilot Chat.

Why is it gaining traction?

Unlike static diff tools or verbose prompt engineering, crit delivers a familiar PR review flow in a single binary, with live file watching that resets comments on agent edits for seamless loops. Developers love the "Finish Review" button that copies a ready-to-paste prompt, plus suggestion mode for quick edits—streamlining feedback without leaving the browser. It embeds everything needed, no deps, making it instant for GitHub repo workflows or agent's output reviews.

Who should use this?

Prompt engineers and AI coding teams iterating on plans with GitHub Copilot Chat or custom agents. Backend devs planning critical infrastructure technologies via markdown, or anyone tired of copy-pasting feedback into chats. Ideal for multi-round refinement where agents update files based on your crit .review.md.

Verdict

Grab it via Homebrew or binaries if you're building with LLM agents—docs are crisp, tests thorough, workflow addictive despite 15 stars and 1.0% credibility score. Still early; monitor for critical mass before production pipelines.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.