Marketplace›Web and tech›Website development›Run a pre-launch security check on your Lovable, Base44, Bolt or Supabase app, with a pass or fail report
Run a pre-launch security check on your Lovable, Base44, Bolt or Supabase app, with a pass or fail report

Run a pre-launch security check on your Lovable, Base44, Bolt or Supabase app, with a pass or fail report

Website development · delivered in 1-3 days

Within 1 to 3 days, we run a fixed script against your live or staging Lovable, Base44, Bolt or Supabase build and hand you a pass, warn and fail report covering 44 individual checks. This is a scripted audit, not an AI opinion, and it is not a penetration test or certification.

  • Quick scan: the scripted review of headers, HTTPS, exposed files, CORS, client-side secrets, Supabase tables (where used) and a light 20-request rate-limit probe, delivered as a Markdown and HTML report with evidence and plain-worded fixes. Secret values are never printed. One revision included.
  • Scan + fix list: everything above plus a prioritised, plain-language action list written for Lovable, Base44, Bolt or Supabase projects, telling you what to change and where, with a note on anything that is simply a platform default rather than a real issue. One revision included.
  • Scan + fixes: everything above plus us applying the safe fixes ourselves, headers, exposed files and source maps, Row Level Security, and stripping leaked keys from the client, then re-running the script and sending before-and-after reports. Any key rotation in providers such as Stripe or Supabase is guided and only done with your approval. Larger rebuilds or a full code review are quoted on their own. One revision included.

Many Lovable, Base44, Bolt or Supabase builds look complete in the browser while quietly shipping a missing header, an exposed .env file, a secret key sat inside the JavaScript, or a database table anyone can query. These gaps rarely show up until someone goes looking.

We point a fixed script at your live or staging address and return a plain pass, warn and fail report, backed by evidence and a fix described in plain words. The result comes from the script, not from an AI forming an opinion.

What the check covers:

  • Security headers (covering content security, transport security, framing, content sniffing, referrer and permissions policies) and whether HTTP redirects to HTTPS
  • Exposed files such as .env, .git and backups, plus any public JavaScript source maps
  • CORS settings and a scan of public JavaScript for secret keys, reporting only the key type and its first 4 characters, never the full value
  • For Supabase builds: a read-only list of tables reachable with the public anon key, flagging any table that hands back a row to an anonymous visitor (a single-row check only, no writes)
  • A light rate-limit probe of up to 20 requests against one endpoint, not a load test
  • A combined Markdown and HTML report giving pass, warn and fail status, the evidence, and the fix in plain words

What we ask from you:

  • The domain this order is for, live or staging
  • Where Supabase is used and the public key is not already visible on the page, just the project URL and the anon or publishable key, never the service_role or secret key
  • Which site to review, if you run more than one

How the order runs: you share the brief and the URL, we run the script and check the output. With Scan + fix list you also receive an ordered list of changes. With Scan + fixes we handle the safe repairs, re-run the script, and send you the before and after reports.

Handled entirely remotely, with no calls required; questions and updates run through the order messages.

Priced upfront. Should anything not match what was agreed, we put it right at no charge within 7 days of handover, with the clock starting once we have the URL and any key needed.

To be clear about scope: this is a scripted pre-launch review, not a penetration test, not a certification, and not a promise that the app is secure. It does not touch social engineering, paid third-party scanning tools, or a full source-code review; that work is quoted separately so the cost is known in advance.

What happens after you buy

Your order opens its own thread here the moment it is paid, and everything about that order - questions, changes and the final report - happens in it.

Questions people ask

Is this a penetration test or a certification?

No. This is a scripted pre-launch review rather than a penetration test or certification, and it does not guarantee the app is secure. That kind of work is priced separately.

Do you need my service_role or secret key?

Never. Where Supabase is used and the public key is not already on the page, we only need the project URL together with the anon or publishable key.

Where do questions and updates happen?

Everything happens on the private job-status page tied to your order, reached by a secret link rather than a shared portal. Questions, updates and the finished report all live there.

Will you change my app?

Quick scan and Scan + fix list only read what is already public on your site. With Scan + fixes we carry out the safe repairs ourselves, and you sign off on any key rotation or revocation before it happens.

What if something is wrong with the result?

If the outcome does not match what we agreed, we correct it at no cost within 7 days of handover.

People also buy

Website development

AI assistant site readability check

A one-off report checking whether AI assistants and crawlers can find, read and understand your site, with a prioritised fix list delivered in 2 business days.

$98.99PER ORDER 2 days
Website development

AI-built app fixes and completion

We diagnose and repair apps built on Lovable, Bolt, Cursor or Base44, delivering a working fix or finished build within 3 business days.

$223.99PER ORDER 3 days