Security scanner for Cursor-built apps
Cursor helps you write the code; you deploy it yourself. This scanner checks the result: the live, deployed app. It looks for the mistakes that survive into production, like a key that made it into the browser bundle or a database that answers anonymous requests.
Scan your deployed app
What it means
A value shaped like a private key or password is in code your site serves, where anyone can read it. If it still works, anyone who copies it can act as you: run up bills, read your data, or send email in your name. xlogs did not test it.
Why AI tools cause it
AI coding tools often paste an API key straight into a file to make something work, instead of hiding it in a private environment variable. It runs fine, so nothing warns you.
How xlogs checks it
xlogs scans your live JavaScript against 18 key formats covering Stripe (live, test and webhook secrets), OpenAI, Anthropic, AWS, Google, GitHub, Slack, private key blocks, and Supabase server keys (service_role and sb_secret_).
What to check on a Cursor-built app that is live
The fastest path to a working feature is not always the safe one, and AI-written code sometimes takes it: a secret pasted where the browser can read it, a database call with no access rules behind it, or a build that ships source maps. Those are exactly the things a scan of the deployed app can catch.
This checks your live URL only. Source-level issues like SQL injection or unsafe HTML are found by reading the code, not the deployed site, so they are a separate job from this live scan.
The top issues to check on a deployed Cursor app
- An API key or service key exposed in the client bundle
- A database table that returns rows to anonymous requests
- Missing security headers (CSP, HSTS, X-Frame-Options)
- Public source maps or private files served from the live site
How to fix it
The goal: The exposed key has been replaced and revoked at the provider, because removing it from the code does not stop it working, and the new value is stored only in a server-side secret or environment variable, never in code or the client bundle.
- Contain and rotate the exposed key first, because removing it from the code does not stop it working: if you suspect it is being misused, revoke it at the provider now; otherwise create a replacement, put it in the app's server-side secret settings, confirm the app works with it, and then revoke the exposed key promptly
- Remove the literal value from the code and any client-side bundle (the affected spot), so the key is read only from a server-only environment variable
- Make sure the .env file is in .gitignore
- Clear the old value from deployed builds and caches, and from the repository's history where you can; once the key is revoked, a copy left there no longer works
In my app (the spot the scan shows): a credential-shaped value is publicly served. Please make this true: The exposed key has been replaced and revoked at the provider, because removing it from the code does not stop it working, and the new value is stored only in a server-side secret or environment variable, never in code or the client bundle. Steps: Contain and rotate the exposed key first, because removing it from the code does not stop it working: if you suspect it is being misused, revoke it at the provider now; otherwise create a replacement, put it in the app's server-side secret settings, confirm the app works with it, and then revoke the exposed key promptly; Remove the literal value from the code and any client-side bundle (the spot the scan shows), so the key is read only from a server-only environment variable; Make sure the .env file is in .gitignore; Clear the old value from deployed builds and caches, and from the repository's history where you can; once the key is revoked, a copy left there no longer works. Then tell me exactly what you changed, and do not print any secret values back to me or write one into a commit message, a pull request description or a log.
Then verify: After you deploy the fix, xlogs re-scans your live app and reports whether the key is still found in what it reads. That cannot show the old key was revoked, so revoke it at the provider either way.
Full step-by-step fix guide, with a copy-paste block for each AI tool →
Common questions
Can xlogs scan my Cursor project directly?
Not the local project. This web scanner checks a deployed URL, so point it at your live app once it is online. It finds the issues that reach production: exposed keys, an open database, missing headers, and public source. Source-only issues like SQL injection are a separate, code-reading check.
Why check the live app if I can read the code in Cursor?
Some problems only appear once the app is built and deployed: which secrets ended up in the browser bundle, whether source maps shipped, and how your database actually responds to an anonymous request. The live scan sees exactly what a visitor (or an attacker) sees.