New AI-assisted pull request fixes

GitHub security scanner
for AI-built apps.

Review your GitHub repository before launch. Check exposed secrets, unsafe database rules and npm dependencies, then use an optional AI review for authentication and payment risks. Turn findings into fix prompts or pull requests you review.

3 free scans and 2 free fixes. Pro from £10 a month.

No code stored You approve every change
Built for teams shipping withCursorLovableClaudeChatGPTGitHub
Product walkthrough

See Quick Scan First in action.

From opening a repository to reviewing the findings and preparing a fix.

Review before release

From code findings to a pull request you control.

Start with built-in checks, then choose deeper review when your launch needs it.

Built-in checks
  • Check selected repository files for known risky patterns
  • Find committed .env files, possible secrets and unsafe database rules
  • Check supported npm lockfiles against OSV.dev
Workflow

From repository to reviewed fix.

A focused release check that fits the way you already work.

01

Open a repository

Quick Scan First reads the GitHub repository you choose using a fine-grained token you control.

02

Run a focused review

Static checks and AI review look for secrets, unsafe permissions, broken payments and production risks.

03

Ship a reviewed fix

Generate a pull request, inspect the changes on GitHub, and merge only when you are satisfied.

Quick Scan First robot repairing a vulnerable code pathway
Automated remediationA reviewed fix, ready for your pull request.
Coverage

The release risks that are easy to miss.

Fast deterministic checks for known mistakes, with deeper reasoning when you need it.

Private by design.
Your source is processed for the scan, not stored as a copy.

Repository-wide coverageFind the risk hiding between files.

Included with every scan

  • Committed environment files
  • Hardcoded API keys and private keys
  • Open Firebase and Supabase access
  • Known npm dependency vulnerabilities
  • Missing repository safeguards

Deep review with Pro or credits

AI
  • Missing authentication and ownership checks
  • Payment and webhook implementation mistakes
  • Crashes across important user paths
  • Data exposure and broken configuration
  • Evidence linked directly to your source code
Pricing

Start with the basics. Upgrade when you need depth.

Use the free checks, subscribe for regular reviews, or buy credits when a launch needs extra attention.

Free
£0forever

For getting the fundamentals right.

  • 3 built-in scans
  • 2 one-click fixes
  • Ready-made fix prompts
Add to Chrome
Credits
Pay once

Extra capacity without a subscription.

25 credits£7
60 credits£15
150 credits£30
AI scan: 3 credits · Fix: 5 creditsThe price you see is the price you pay.
Questions

The details, clearly answered.

How access, code handling, AI reviews and billing work.

Will it change my code without asking?+

No. A fix is always a pull request on a new branch. Nothing changes until you review and merge it. Quick Scan First will not touch workflows or anything outside your repository.

What access does it need?+

A fine-grained GitHub token, limited to repositories you choose, with read and write access to Contents and Pull requests. You can revoke it at any time.

Is my source code stored?+

No. Code is read during a scan and is not saved on our servers. Relevant files are sent to Anthropic only when an AI review or fix needs them.

Does it work with private repositories?+

Yes, provided your fine-grained token has access to them.

Can the AI be wrong?+

Yes. Quick Scan First is a useful second pair of eyes, not a security audit or guarantee. Always review generated pull requests before merging.

What is a credit?+

Credits are pay-as-you-go. An AI scan costs 3 credits and a fix costs 5. They do not expire, and failed operations are refunded.

Is VAT added at checkout?+

No. The price you see is the price you pay, with nothing added on top at checkout.

Can I cancel Pro?+

Yes. Cancel from the billing page in the extension and keep Pro access until the end of the paid period.

Choose your review

Explore the checks behind your launch.

Find the workflow that fits your repository and the risks you need to investigate.

Product guide

AI code security scanner for vibe-coded apps

Review GitHub code built with Cursor, Lovable, Claude or ChatGPT for exposed secrets, unsafe database rules and launch risks. Keep control of every proposed fix.

Explore the check →
Product guide

Supabase RLS and Firebase rules checks before launch

Check repository SQL and Firebase rules for unsafe access patterns before launch. Understand what source checks reveal and what to verify in your live database.

Explore the check →
Product guide

npm dependency vulnerability scanner for GitHub

Check supported npm package-lock files against OSV.dev for known vulnerabilities before launch. Review affected packages, fixes and the limits of a dependency scan.

Explore the check →
Product guide

Stripe webhook and payment code review before launch

Use AI-assisted repository review to investigate Stripe webhook signatures, server-side payment checks and ownership mistakes before launching a paid app.

Explore the check →
Buying guide

Compare GitHub code security tools for pre-launch review

Compare secret scanning, SAST, dependency checks and AI code review workflows. See where Quick Scan First fits without assuming one scanner replaces every tool.

Explore the check →
Product guide

GitHub repository security scanner for pre-launch checks

Scan a GitHub repository for exposed secrets, unsafe database rules, vulnerable dependencies, missing authentication checks and payment risks before launch.

Explore the check →
Product guide

Pre-launch code review for web apps

Use a repeatable pre-launch code review to find secrets, broken access control, payment mistakes, dependency risks and production configuration problems.

Explore the check →
Before you ship

Give your code one more pair of eyes.

Add to Chrome
Quick Scan First robot confirming that a repository is ready to launch
Ready signalShip with confidence.