Base44 Security: RLS, Roles and Mistakes to Avoid
How Base44 security works: row-level and field-level rules, admin roles, service role risks, the 2025 Wiz finding and how to test your app before launch.
- Base44 gives you server-side row-level security (RLS) and field-level security (FLS) on every entity. They only protect you if you set them.
- Hiding a button or a page is not security. The rules on the data are what count.
- Backend functions that use the service role skip RLS and FLS completely, so guard who can call them.
- Test as a logged-out visitor and as a second ordinary user before launch.
Base44 is secure enough for real business apps if you set up its security rules correctly. The platform handles login, HTTPS and server-side permission checks. Your job is to set row-level and field-level security on every entity so each user can only see and change what they should. Most Base44 security problems we see come from missing or loose rules, not from the platform itself.
We are a Base44 Legend Partner and review Base44 apps for security as part of client work. We test rules against the live database as an anonymous visitor through the API, not only through the app's screens. This guide covers what to set, what to avoid and how to test.
Is Base44 secure?
The platform is built with the right tools: authentication with email and password, social logins and custom SSO, server-enforced data rules, and automatic HTTPS on custom domains, per the Base44 backend docs. Like any platform, it has had issues. In July 2025, the research firm Wiz found a login bypass, and Base44 fixed it within a day. More on that below.
The bigger risk is the app you build. AI-written code in general often has security gaps.
What is row-level security in Base44?
Row-level security (RLS) decides which records each user can create, read, update or delete. You set a rule for each of those four actions on each entity. A rule can allow everyone, block everyone, or allow only records that match a condition, as the Base44 security docs describe.
Conditions can check:
- The user's role, for example only admins.
- A match between the user and the record, for example the record's creator email equals the logged-in user's email.
- Fields on the record itself, for example only records marked as published.
- Combinations, using operators for "or", "and", "not" and "in a list".
A good default for most business data is: users can read and edit only their own records, and admins can do everything.
What happens when Base44 denies access?
Base44 is careful not to leak information when it blocks a request.
Denied reads behave as if the data doesn't exist, so they don't reveal what's there.
In practice, a blocked list or filter returns an empty result, a blocked lookup of a single record returns not found, and a blocked create or update returns a forbidden error. That is useful to know when testing: an empty list is what a working rule looks like.
What is field-level security in Base44?
Field-level security (FLS) hides or locks individual fields inside a record. The Base44 docs use a salary field as the example: everyone on a team can see a staff record, but only admins can see the salary. Use FLS for internal notes, pricing, cost data, personal details and anything a user should not be able to change, like their own role.
What is the service role and why is it risky?
Inside backend functions, Base44 lets code act with full admin rights through the service role. That is useful for jobs like nightly reports or processing a payment, but it removes every protection you set.
bypasses both RLS and FLS entirely, regardless of how the entity is configured
If a backend function uses the service role, check who can call it and what input it accepts. A function that takes a record ID from the user and returns that record with admin rights can expose your whole database.
What are the most common Base44 security mistakes?
- 1Leaving entities open. An entity with no read rule, or a rule that allows everyone, can be read by anyone who finds the API.
- 2Relying on hidden screens. Hiding the admin page from regular users does not stop them from requesting admin data directly.
- 3Letting users edit their own role. If the role field is writable, any user can make themselves an admin. Lock it with FLS.
- 4Matching on the wrong field. A rule that matches a field the user can edit, such as an owner email they can change, can be bypassed.
- 5Overusing the service role. Every backend function with admin rights is a door that skips your rules.
- 6Forgetting logged-out visitors. Rules that reference the user block anonymous callers, but rules set to allow everyone do not.
- 7Putting API keys in the front end. Keys belong in secrets used by backend functions, never in browser code.
How do you test the security of a Base44 app?
Test before launch and again after any big change. Here is the process we follow:
- 1List every entity and write down who should be able to create, read, update and delete each one.
- 2Compare that list with the actual RLS and FLS rules in the app. Fix any gaps.
- 3Log out and try to load data. As an anonymous visitor, private entities should return nothing.
- 4Create two ordinary user accounts. As user A, try to open, edit and delete user B's records by changing IDs in the address bar.
- 5Try to change protected fields, such as your own role, and confirm the change is refused.
- 6Review every backend function that uses the service role and check what it trusts from the caller.
What was the 2025 Base44 vulnerability?
In July 2025, Wiz found that undocumented registration and verification endpoints let someone create an account on a private Base44 app using only its public app ID. That could bypass login, including on apps protected by SSO.
Wiz reported it on July 9, Base44 fixed it on July 10, and Wiz published its findings on July 29, 2025. Coverage in The Hacker News and the Wiz write-up both note the quick fix.
Fortunately, Wix confirmed that there was no evidence of exploitation in the wild.
The lesson for builders: platform bugs happen on every platform, and good RLS limits the damage. An attacker who got past login on an app with strict rules would still only see what an ordinary new user could see.
Should you get a Base44 security review?
If your app stores customer data, payments, health or financial information, or anything you would not want public, yes. A review is quick compared with the cost of a leak. We are a Base44 Legend Partner and Base44 Ambassador, and security checks are part of every app we ship. See our Base44 developers page or app development services.
Building your first app? Our Base44 tutorial sets up RLS from the start, and our Base44 integrations guide covers keeping API keys safe.
Frequently asked questions
Is Base44 safe to use for business apps?
Yes, if you set row-level and field-level security on every entity and test them. The platform provides login, HTTPS and server-side rules, but it cannot guess who should see your data.
Does Base44 have row-level security?
Yes. You can set separate rules for create, read, update and delete on each entity, based on user role, a match between the user and the record, or fields on the record.
Was Base44 hacked?
In July 2025, Wiz found a login bypass in Base44 and reported it privately. It was fixed the next day, and Wix said there was no evidence it was exploited.
Can users see each other's data in Base44?
Only if your rules allow it. Without a read rule that limits records to their owner, any logged-in user may be able to read every record. Set RLS and test with two accounts.
What does the Base44 service role do?
It lets backend functions act with full admin rights, skipping all RLS and FLS rules. Use it only where needed and check who can call those functions.
Is AI-generated code secure?
Not by default. Veracode found security flaws in 45% of coding tasks across more than 100 AI models in 2025. Review permissions and test before launch, whatever builder you use.
Roenic runs Virtual Horizons and is a Base44 Legend Partner, Ambassador and Community Pillar, Base44 Shark Tank winner and multiple-time Vibe Sprint winner. The studio builds web and mobile apps, AI voice agents and automations, with more than 10 apps live on the App Store and Google Play. More about us
Sources
- 1.Base44 documentation: Entity security
- 2.Base44 documentation: Backend features
- 3.Wiz Research: Critical vulnerability in Base44
- 4.The Hacker News: Wiz uncovers critical access bypass flaw in AI-powered vibe coding platform Base44
- 5.Veracode: AI-generated code poses major security risks in nearly half of all development tasks
Keep reading
Base44 Review 2026: An Honest Take From a Base44 Partner
An honest Base44 review from a Base44 Legend Partner with 10+ apps live in the app stores: what it does well, where it struggles, and who should use it.
Read the guideWhat Is Base44? A Plain-English Guide for 2026
Base44 is an AI app builder that turns a written description into a working web app with a database, logins and hosting. How it works, and who it suits.
Read the guideBase44 Pricing 2026: Plans, Credits and Real Total Cost
Base44 pricing explained: every plan, what credits mean, which plan you need for a custom domain or the App Store, and the costs most guides leave out.
Read the guide