For apps built with AI tools

From prototype to production.

You built your app with Lovable, Bolt, Replit or Cursor, and it works. Before real users and real payments arrive, have an engineer check the parts AI tools often get wrong: secrets, database access, login and payments. You keep the code.

  • Fixed-price review
  • Written report, fixes ranked by risk
  • Your repo, your code
Sample findings
What we check

The parts that break when real users arrive

AI tools are good at screens and flows. They tend to make the same mistakes in the parts you can't see, so we check those against the same list every time.

Secrets and keys

API keys, service tokens and passwords that ended up in the browser, the repo or a public log.

Database access

Access rules on every table, so each user reads and changes only their own data.

Login and accounts

Sign-up, password reset, sessions and admin roles, checked on the server, not just hidden in the interface.

Payments and paywalls

Checkout, webhooks, refunds and paid features, so nobody gets in free and nobody gets charged twice.

Deploys and environments

Separate test and live setups, safe database changes, and a deploy you can roll back.

Monitoring and backups

Error alerts, uptime checks and backups you have actually restored from once.

How it works

A review first, then only the fixes you choose

You see what's wrong and what it costs to fix before you commit to any build work.

  1. 01

    Send the app

    Read-only access to your code and a link to the app. We agree on a fixed price for the review.

  2. 02

    Review

    We go through the checklist and write a plain-English report, with every fix ranked by risk.

  3. 03

    Walkthrough call

    We go through the report together, answer questions and quote the fixes at a fixed price.

  4. 04

    Fixes, if you want them

    We make the changes in your repo, with tests on the risky paths. Or hand the report to your own developer.

Pricing

Fixed prices, agreed before we start

Every job is quoted up front. If the work turns out bigger than the quote, we tell you before we do it.

Readiness review

Start here

Find out what needs fixing before launch.

From $750fixed price, set by the size of the app
  • Secrets, database access, login, payments, deploys and monitoring
  • Written report with fixes ranked by risk
  • Walkthrough call
  • Fixed quote for the fixes
Book a review

Fix sprint

The fixes from your report, done in your code.

Fixed quotepriced from the review report
  • Every fix you choose from the report
  • Tests on the risky paths
  • Changes in your repo, explained
  • Safe deploy to production
Ask about fixes

Production care

Keep it running safely after launch.

From $150a month, month to month
  • Error and uptime monitoring
  • Dependency and security updates
  • Backups we test
  • A set number of small changes each month
Ask about care
How we work

Your app, made safe. Not rewritten.

We fix what you built rather than starting again, and we explain every change so you, or the next developer, can keep going without us.

  • You keep the code

    Everything lives in your repo and your accounts. No lock-in, no proprietary layer.

  • Plain-English reports

    Each finding says what's wrong, what could happen, and how to fix it.

  • Least access

    Read-only access for the review, and access you can revoke for the fixes.

  • Keep building with AI

    We leave notes your AI tool can follow, so new features don't undo the fixes.

Questions

Straight answers

Do I need to be technical?

No. The report is written for founders, and the walkthrough call covers anything that isn't clear. Your developer, if you have one, gets the technical detail too.

Which tools do you work with?

Apps from Lovable, Bolt, Replit, Cursor and v0, and hand-written code too. Common stacks include React and Next.js with Supabase or Firebase, and Stripe for payments. If yours is different, ask.

What access do you need?

For the review, read-only access to your code and a link to the running app. For fixes, access to the repo and hosting that you can remove when we're done.

Is a review a guarantee that my app is secure?

No review can promise that. It lowers your risk a great deal, and the report says exactly what we checked, so you know what is covered and what isn't.

Will you rewrite my app?

Only where a part can't be made safe as it is. We fix what you built, and the report says up front if any part needs replacing.

Can I keep building with my AI tool afterwards?

Yes. We leave notes in the repo that describe the rules the code now follows, so you and your AI tool can keep adding features without undoing the fixes.

Book a review

Tell us about your app

We'll reply with a fixed price for the review and when we can start.

This opens your email app with your answers filled in.