✓ Copied to clipboard
Commands Scoring Vision Pricing Install Pricing ↗
AI-Powered CLI · Zero Config

Stop shipping
tech debt.

Ratchet scans every file, scores every weakness, and commits AI fixes that pass your tests — one relentless click at a time.

Node.js ≥ 18 required
6
quality dimensions
100
point scoring system
0
regressions allowed
AI
powered fixes
ratchet — ~/projects/my-app
ratchet torque --target error-handling --clicks 7 Target : error-handling Path : src/api/ Clicks : 7 Tests : npm test Score: 62/100 (14 issues found) Targeting: Empty catches (5/6), any types (4/5), functions >50 lines (3/4) Click 1 — passed [a3f9b21] — Score: 62 → 65 (+3) — 2 issues fixed Click 2 — rolled back Click 3 — passed [7bc1d44] — Score: 65 → 68 (+3) — 1 issue fixed Click 4 — passed [2e8f053] — Score: 68 → 71 (+3) Click 5 — passed [9da3c17] — Score: 71 → 73 (+2) — 1 issue fixed Click 6 — rolled back Click 7 — passed [f81b44a] — Score: 73 → 76 (+3) — 2 issues fixed Done. 5 landed · 2 rolled back · 4m 12s
How It Works

Up and running in minutes.

Three commands are all it takes to go from zero to a cleaner codebase.

Step 01
📦
Install
One global install. Works with any Node.js project — no config files, no setup ceremony.
npm install -g ratchet-run
Step 02
🔍
Scan
Get an instant quality score across 6 dimensions — security, testing, type safety, performance, and more.
ratchet scan
Step 03
Improve
Let AI fix what it finds. Every change runs your test suite — only improvements that pass get committed.
ratchet torque --clicks 10
The Mechanism

Built like a ratchet wrench.

A ratchet only turns one way. Each click advances the socket — it can never slip back. Ratchet works the same way on your code.

Click
One full cycle
Analyze → propose → build → test → commit. One click, one real improvement.
🔩
Torque
The driving force
The command that applies sustained force to your codebase over multiple clicks.
🛡
The Pawl
Anti-rollback
If tests fail, the change is reverted automatically. Nothing bad ever commits.
🚀
Tighten
Ship it
Finalize the run and open a pull request with all the clean, tested improvements.
Each click cycle
Analyze
Propose
Build
Test
Commit ✓
repeat
Four Commands

Everything you need.
Nothing you don't.

A focused toolkit. Each command does one thing exceptionally well.

ratchet scan
Know exactly where you stand
🔍

Scores every file in your codebase across six quality dimensions. Produces a ranked list of issues with context, so you always know what to fix next.

ratchet scan --path src/
  • Per-file and aggregate scores out of 100
  • Issue breakdown with line-level context
  • Exportable as JSON, Markdown, or PDF
  • Zero config — auto-detects project type
ratchet torque
Set it and let it run
🔄

Spawns an AI agent that iteratively improves a target area across as many clicks as you give it. Each passing click commits clean, tested code.

ratchet torque --target security --clicks 10
  • Continuous analyze → fix → test → commit loop
  • The Pawl rolls back anything that breaks tests
  • Re-scores after each successful click
  • Works across any target: files, dirs, or themes
ratchet improve
Surgical, focused fixes

Like torque but more targeted — point it at a specific file or issue category for concentrated, high-precision improvements without broad sweeps.

ratchet improve --file src/auth/middleware.ts
  • Single-file or single-issue focus
  • Faster iteration on specific problem areas
  • Ideal for PR-level clean-up before shipping
  • Same safety guarantees as torque
ratchet vision
A living map of your codebase
🌐

Generates an interactive dependency graph — nodes color-coded by health, sized by blast radius. Instantly see which files are risky and which need attention.

ratchet vision --open
  • Self-contained HTML, no server needed
  • Nodes sized by number of dependents
  • Color-coded: green (healthy) → red (critical)
  • Zoom, filter, and click to explore any node
What developers say

Used by engineers who care about quality.

★★★★★
"

Ratchet took our codebase from a 58 to a 79 in one afternoon. I pointed it at our auth module, walked away for an hour, and came back to 12 merged commits — all green.

👩‍💻
Sarah K.
Senior Engineer, Series B SaaS
★★★★★
"

Finally, a tool that doesn't just tell you what's wrong — it actually fixes it. The Pawl mechanism is brilliant. I've never seen a bad commit land from a torque run.

👨‍💻
Marcus T.
Tech Lead, Fintech Startup
★★★★★
"

We run ratchet torque on every PR branch. Code review time dropped by 40% because the obvious stuff is already fixed before a human looks at it.

👩‍🔬
Priya M.
Engineering Manager, Scale-up
my-api/
Production Readiness Score
76/100
Code Quality 18/24
Testing 16/20
Security 10/16
Error Handling 12/14
Performance 11/14
Type Safety 9/12
ratchet scan

A number for every file.
Not just a vibe.

Stop guessing what "clean code" means on your project. Ratchet produces an objective, reproducible quality score across every dimension that matters for production systems.

Run it before a sprint to find your biggest risks. Run it after torque to measure progress. Track it over time to prove your codebase is improving.

Code Quality
Testing
Security
Error Handling
Performance
Type Safety
ratchet vision

A living map of
your codebase — color-coded by health,
sized by risk.

Run ratchet vision and get a self-contained HTML file: an interactive force-directed graph of every file and dependency in your project.

Nodes glow green when healthy, amber when there are issues, red when critical. Node size reflects blast radius — the larger it is, the more your codebase depends on it.

Green — Score ≥ 80. Healthy, well-tested, low risk.
Amber — Score 50–79. Issues present, improvement needed.
Red — Score < 50. Critical attention required.

Generated as a single HTML file. No server, no npm, no build step. Open it in any browser.

index.ts auth.ts router.ts db.ts utils.ts config.ts Score: 76/100 14 files · 3 critical Healthy ≥80 Issues 50–79 Critical
ratchet vision — interactive graph
The Pawl

It can only
move forward.

The Pawl is Ratchet's anti-rollback mechanism. Every proposed change runs your full test suite before it commits. If tests fail, the change is automatically reverted and the click doesn't count.

The result: a branch of commits that only ever moves forward. Every merged click is a proven improvement — no exceptions.

Start improving →
Click 1 — removed empty catches [a3f9b21]
+3
Click 2 — tests failed · rolled back
−0
Click 3 — typed API responses [7bc1d44]
+3
Click 4 — split 80-line function [2e8f053]
+3
Click 5 — tests failed · rolled back
−0
Click 6 — added structured logging [9da3c17]
+2
Click 7 — stricter null checks [f81b44a]
+3
5 landed  ·  2 rolled back  ·  Score: 62 → 76 (+14)
Pricing

Simple, honest pricing.

Start free. Level up when your codebase demands it.

Monthly Annual  20% off
Free
$0
 
See your score. Feel the itch.
  • Scan: unlimited
  • Vision: 1/week
  • Quality score across 6 dimensions
  • JSON & Markdown export
Get Started
Builder
$12
/mo
BYOK. Your key, your rate.
  • 30 improvement cycles/month
  • BYOK — bring your own API key
  • ratchet improve access
  • Everything in Free
Start Building
Team
$149
/mo
Hosted. No API key needed.
  • 5 seats included
  • 500 improvement cycles/month
  • Hosted — no API key required
  • CI/CD GitHub Action
  • Slack & Discord notifications
  • Team dashboard + score trends
Start Team Trial
Enterprise
Custom
 
Air-gapped. Your rules. Your SLA.
  • Unlimited cycles
  • Air-gapped / self-hosted
  • Custom scoring rules
  • SLA + dedicated support
  • Everything in Team
Contact Us

Need more cycles? $0.75/cycle on any paid plan.

Get Started

Ship better code,
every single day.

Install Ratchet globally and point it at any project. No configuration required to start.

1
Install
npm install -g ratchet-run
2
Scan your codebase
ratchet scan
3
Start improving
ratchet torque --clicks 5
4
Open a PR with all improvements
ratchet tighten --pr

Prerequisites: Node.js ≥ 18, git, and an AI coding agent on your PATH.

View Pricing See a demo →
FAQ

Common questions.

Everything you need to know before you run your first torque.

Does Ratchet ever break my code?
No. That's the whole point of the Pawl. Every AI-proposed change runs your full test suite before it commits. If a single test fails, the change is automatically reverted and the click is discarded — nothing bad ever lands in your repo.
What languages and frameworks does Ratchet support?
Ratchet works with any project that has a test command. It auto-detects JavaScript, TypeScript, Python, Go, Rust, Ruby, and Java projects. As long as you have a test script (or specify one via --test-cmd), Ratchet can improve it.
What AI model does Ratchet use?
On Builder and Pro plans, Ratchet uses your own API key (BYOK) with Claude or GPT-4 — you control the model and the cost. Team and Enterprise plans include hosted AI with no API key required.
What counts as one "improvement cycle"?
One click = one full analyze → propose → build → test cycle. If the change passes tests and commits, that's one cycle used. If tests fail and the change rolls back, it doesn't count against your limit. You're only charged for successful improvements.
Can I run Ratchet in CI/CD?
Yes. Ratchet ships with a GitHub Action on Team and Enterprise plans. You can configure it to run torque on every PR, on a schedule, or against specific files changed in a diff. The action opens a follow-up PR with all landed improvements.
Is my code sent to a third party?
On BYOK plans, your code goes directly to your chosen AI provider (Anthropic or OpenAI) — never through Ratchet's servers. Enterprise customers can deploy Ratchet fully air-gapped with their own self-hosted models.