Crit
Line-level review for AI coding agents
Crit solves a real pain: reviewing agent output in the terminal is terrible. Its local-first design and broad agent compatibility make it a practical choice for anyone iterating with AI. The project is actively maintained and free—worth installing today.
- Solo developers using AI coding agents (Claude Code, Cursor, Copilot, etc.)
- Teams iterating on agent-generated code with local-first review
- Developers who want PR-style review without leaving local environment
- Anyone tired of reading terminal diffs and missing details in agent output
- Non-developers or pure no-code users
- Teams requiring built-in user management or RBAC
- Developers who prefer terminal-only workflows
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
In short
Crit — Line-level review for AI coding agents. Best for Solo developers using AI coding agents (Claude Code, Cursor, Copilot, etc.), Teams iterating on agent-generated code with local-first review, Developers who want PR-style review without leaving local environment. Free to use.
What's new in Crit
Checked 11 days agoAcross the latest 2 updates: 2 changelog entries.
Crit v0.17.1: Chrome DevTools cookies in crit live, range mode working changes, stack popover orientation
Live mode can pull upstream session cookies from Chrome debugging. Range mode adds virtual working changes entry and stack orientation cues.
Crit v0.17.0: User-defined session finish hooks and --session flag
Users can define finish prompt hooks for review rounds. New --session flag enables cwd-independent session reconnect.
Viability Score
How likely is Crit to still be operational in 12 months? Based on 4 signals — momentum (how recently it shipped), wrapper dependency, revenue model, and web presence.
Last calculated: July 2026
How we score →Key Features
- Inline comments on plans, code diffs, live apps, and static HTML
- Git diff and round-by-round diff views
- AI review loop: comment, hand off to agent, iterate
- Vim keybindings for keyboard-driven workflow
- One-click shareable reviews for async collaboration
- Syntax highlighting for 190+ languages
- Mermaid diagram rendering from fenced blocks
- Four review modes: markdown/plans, branch/PR diffs, live app proxy, static HTML iframe
- Session finish hooks for custom agent instructions on round completion (v0.17.0)
- Compare-against picker and commit range controls
- Chrome DevTools cookie sharing for live mode (v0.17.1)
- Local-first, no data leaves your machine unless you share
- Supports stacked PRs (Git, jj, sapling)
- Comment on DOM elements in live apps (automatic reload)
- Virtual working-change entries in commit picker (v0.17.1)
About Crit
Crit is a local-first review tool for developers working with AI coding agents. Instead of scrolling through terminal diffs, Crit opens agent changes in a browser where you can comment on any line, section, or element. It supports four review modes: markdown/plans, branch/PR diffs, live app proxy, and static HTML iframe. Whether your agent wrote a plan, modified a branch, launched a dev server, or generated an HTML artifact, Crit provides a dedicated review surface with syntax highlighting for 190+ languages, Mermaid diagram rendering, and round-by-round diff views. Every review produces a structured file that any file-reading agent can consume, with no API or plugin framework needed. Crit integrates directly with Claude Code, Cursor, GitHub Copilot, OpenCode, Codex, Gemini CLI, Qwen Code, Hermes, Pi, Grok, Windsurf, Cline, and Aider. The tool is open source and self-hostable, emphasizing privacy: reviews stay on your machine unless you explicitly share them. Recent updates (v0.17.1, July 2026) introduced Chrome DevTools cookie sharing for live mode, virtual working-change entries in the commit picker, and session finish hooks for custom agent instructions on round completion. Crit is maintained by Tomasz Tomczyk and has 692 GitHub stars. It's a practical alternative to terminal diffs or generic PR reviews, giving you a persistent, line-attached comment surface that stays across iterations.
Behind the Verdict
We'd reach for Crit when we're deep in a Claude Code or Cursor session and need to review a 300-line plan or a multi-file branch change. The line-level commenting is exactly what's missing from terminal diffs. Crit's four review modes cover the common agent outputs: plans, code diffs, live apps, and static HTML. The setup is straightforward—one command installs a skill or rule for your agent. Crit writes a structured review file that any file-reading agent can consume, so the feedback loop is instant. Where it bites: Crit is local-first, so there's no cloud sync or multi-user workspace. If you need team review workflows or RBAC, you'll need to pair it with git or another tool. The collaborative features are still early. Compared to alternatives like GitHub PR reviews, Crit is faster for local agent output but lacks the team context. Compared to asking the agent to review itself, Crit puts you in the loop—which is the whole point. In practice, Crit shines for solo developers and small teams who want PR-style review without leaving their local environment. It's free and open source, so there's no cost to try. If you're tired of reading terminal diffs and missing details in agent output, give Crit a try—it might change your workflow.
Researching Crit? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Use Cases
- Review a 300-line AI-generated plan by commenting on specific sections in a rendered markdown view.
- Inspect an agent's code changes across 14 files with syntax-highlighted diffs and line-level comments.
- Provide feedback on a running localhost app by tapping on misaligned elements in a live proxy.
- Share an HTML artifact review with a colleague via a one-click public link.
- Iterate with your agent over multiple rounds, seeing exactly what changed each round.
- Automate agent instructions on review approval using custom finish hooks.
Limitations
- Crit is a CLI tool with an optional web UI; it requires command-line comfort and local setup.
- The free plan includes all features, but there is no cloud-hosted version—you must self-host or run locally.
- Agent integration relies on each tool's plugin/skill system, which may lag upstream changes.
- The project is pre-1.0 and may have breaking changes.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Integrations
Resources & Guides
Official links
Tools that pair well with Crit
Common stack mates teams adopt alongside Crit, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Alternatives to Crit
View allFrequently Asked Questions
Categories
Best-of guides
Used Crit? Help shape our editorial sentiment research.