Lovable security scan vs an outside check: what each one sees
Short answer: Lovable's built-in Quick and Deep scans review your project from the inside: database access rules, dependencies and, with the Deep scan, your application code. An outside check such as SniffPup, a free passive security scanner for vibe-coded apps, looks at what your published site sends every visitor. They answer different questions, so run Lovable's scans first, then check the live URL.
Passive, read-only checks only, for a live URL you own or are authorized to test. An outside scan is not a source code or RLS audit, and a score of 100 means no issues were found in our checks, not proof of security. Full fixes free for every critical issue (no critical? one other fix is free).
Home › Guides › Lovable security scan vs an outside check: what each one sees · Last updated · By Ember, SniffPup
What Lovable's built-in scans cover
Quick scan. A fixed set of checks that finishes in seconds and runs automatically every time you publish. It reviews your database access rules and settings (for example tables without row-level security, rules that let everyone through and leaked-password protection turned off), audits your npm dependencies for known vulnerabilities (only critical ones become findings) and checks for an MCP server your app exposes without authentication.
Deep scan. Reviews all your application code and includes everything in the Quick scan. It reports findings on access control, endpoints anyone can trigger, unsafe input and injection, leaked secrets in your source code or repository, payments, sign-in and exposed personal data. It does not run automatically as you work; you start it yourself, and Enterprise workspaces can schedule it.
Optional add-ons. Workspaces can connect Wiz (dependency and source code analysis that runs with Lovable's scans) or Aikido (AI penetration tests against your running app, billed by Aikido). Lovable can also publish a Trust center page at /.well-known/trust.html on an externally published app, with security facts Lovable observes about the live app. Lovable describes it as informational, not a certification.
Lovable is candid about limits: its docs say these tools “do not replace a thorough security review” and “cannot guarantee complete security.”
What an outside check adds
An outside scan doesn't see your project. It sees your published site the way a stranger's browser does. SniffPup's free scan checks the HTML and JavaScript your site ships for secret keys (masked in the report), requests paths such as /.env and /.git/config, and checks source maps, security headers, cookie flags, HTTPS and outdated JavaScript libraries. If your site uses Supabase, it uses only the public anon key your site already ships to send one read-only HEAD request per table, up to 8 tables, to see whether a signed-out visitor can read rows. It never downloads rows and never writes.
What it can't do: read your source code, server logic or RLS policies, log in, or test anything behind sign-in. SniffPup scans live web URLs, and its Mobile app scan checks app builds you export yourself. See what the scanner does and never does.
Side by side
Lovable columns from Lovable's security overview, checked November 10, 2026. “Not listed” means Lovable's docs don't list it for that scan.
| Check | Lovable Quick scan | Lovable Deep scan | Outside passive scan (like SniffPup) |
|---|---|---|---|
| Database access rules (for example tables without row-level security, rules that let everyone through) | Yes | Yes | No. It can't see your policies. With only your public key it tests whether a signed-out visitor can read rows in up to 8 Supabase tables |
| Your application code (access control, endpoints anyone can trigger, injection, payments, sign-in) | No | Yes | No |
| Secrets | Not listed | Yes, in your source code or repository | Only what the live site ships: secret keys in its HTML and JavaScript, masked in the report |
| Dependencies | Yes, known critical npm vulnerabilities | Yes (includes the Quick checks) | Partly: outdated JavaScript libraries the site ships, when the version can be read |
| What the published site serves: public .env or .git files, source maps, security headers, cookie flags, HTTPS | Not listed | Not listed | Yes |
| When it runs | Automatically when you open the publish dialog, or on demand | When you start it (Enterprise workspaces can schedule it) | When you scan the live URL |
A simple routine
1. Run Lovable's Quick scan
Open the publish dialog (the Quick scan starts on its own) or click Run quick scan in the project Security view. It checks your database access rules and settings, known critical npm vulnerabilities and an MCP server exposed without authentication. Fix every Critical finding.
2. Run the Deep scan before you publish
Click Run deep scan in the Security view, or ask Lovable in the project chat to review your app's security. The Deep scan reviews your application code for access control, endpoints anyone can trigger, unsafe input, leaked secrets in source, payments, sign-in and exposed personal data. It does not run automatically as you work.
3. Test your data signed out
Open the app in a private window and confirm a stranger can't read private tables. Review your policies under More, then Database, Manage, Policies. Lovable says to review RLS policies regularly, especially before publishing or after changes to sign-in or the database.
4. Publish, then check the live URL from the outside
Scan the published URL of a site you own. An outside, passive check sees what the deployed site sends every visitor: keys in the JavaScript it ships, public .env or .git files, source maps, headers, cookies and HTTPS. Fix anything new in Lovable, then publish again.
5. Repeat after big changes
Lovable's Quick scan reruns when you open the publish dialog. Run the Deep scan again after significant code changes, and scan the live URL again after each publish. A clean result from any tool means nothing was found in its checks, not that the app is secure.
Next: Is my Lovable app secure?, How to audit a Lovable app, How to hide API keys in Lovable, and the best free security scanners for vibe-coded apps.
FAQ
Is Lovable's built-in security scan enough?
It is the right place to start, and it sees things no outside scanner can, such as your database rules and your application code. Lovable's own docs say the tools do not replace a thorough security review and cannot guarantee complete security. Run the Quick and Deep scans, fix the Critical findings, then check what your published site actually sends to visitors.
What is the difference between Lovable's Quick scan and Deep scan?
Per Lovable's docs, the Quick scan is a fixed set of checks that finishes in seconds and runs automatically when you publish: database access rules and settings, known critical npm dependency vulnerabilities, and an MCP server exposed without authentication. The Deep scan reviews all your application code and includes the Quick checks. It does not run automatically as you work; you start it from the Security view, the Security center or the project chat.
Are Lovable's security scans free?
Lovable's docs (checked November 10, 2026) say running the Quick scan or the Deep scan is free. Try to fix all draws from 10 free fixes shared across your workspaces, and each one becomes available again 24 hours after you use it. A conversational security review in the project chat, and asking Lovable to fix findings while you build, use credits.
Does an outside scan replace Lovable's Deep scan?
No. An outside scan can't read your code, your server logic or your database policies, and it doesn't log in. It answers a different question: what does the live site show any visitor? Use it after Lovable's scans, not instead of them.
What does a SniffPup scan check, and what can't it do?
The free SniffPup scan is passive and read-only, for a live URL you own or are authorized to test. It checks secret keys in your HTML and JavaScript, public .env, .git and source map files, security headers, cookie flags, HTTPS and outdated JavaScript libraries. With only the public anon key your site already ships, it sends one read-only HEAD request per table, up to 8 tables, without downloading rows. It is not a source code or RLS audit, it can't see anything behind a login, and a score of 100 means no issues were found in our checks, not proof of security.
Sources: Lovable: Security overview (Quick and Deep scans, when they run, optional connectors, Trust center and costs), checked November 10, 2026. SniffPup facts are from our scanner page.