Skip to content
Free tool - no signup

Is your Lovable + Supabase app leaking data?

A self-check for apps you own. Find keys that should never ship to the browser, and database tables anyone can read or change. Your full result is shown right here.

Supabase only allows testing your own project, so this tool does not send requests to Supabase's domains, including custom domains that point to them. Part 1 reads your public page; part 2 runs in your own SQL Editor and is analysed in your browser.

Part 1 - about 20 seconds

Public check

We download your app's page and its own JavaScript files, like a browser does, and look for Supabase keys that should never be public. We do not send requests to Supabase's own domains or to your database.

We use the URL only to run this check and log just its host name, to spot abuse. How we handle your data

Part 2 - about 2 minutes

Deep check

The public check cannot see your database rules, so here you run one read-only query in your own project and paste the result. It flags tables without Row Level Security, RLS without policies, policies that allow every row, and public storage buckets.

  1. Open your project in the Supabase dashboard and go to SQL Editor.
  2. Paste the query below and run it. It only reads catalog tables and changes nothing.
  3. Copy the value of the report column and paste it here.
-- Appience Supabase self-check v1 (read-only: reads catalog tables, changes nothing)
with public_tables as (
  select c.oid, c.relname, c.relrowsecurity
  from pg_catalog.pg_class c
  join pg_catalog.pg_namespace n on n.oid = c.relnamespace
  where n.nspname = 'public' and c.relkind in ('r', 'p')
)
select json_build_object(
  'version', 1,
  'tables_without_rls', coalesce((
    select json_agg(t.relname order by t.relname)
    from public_tables t
    where not t.relrowsecurity
  ), '[]'::json),
  'rls_without_policies', coalesce((
    select json_agg(t.relname order by t.relname)
    from public_tables t
    where t.relrowsecurity
      and not exists (select 1 from pg_catalog.pg_policy p where p.polrelid = t.oid)
  ), '[]'::json),
  'permissive_policies', coalesce((
    select json_agg(json_build_object(
      'table', p.tablename, 'policy', p.policyname, 'cmd', p.cmd,
      'roles', p.roles, 'using', p.qual, 'with_check', p.with_check
    ) order by p.tablename, p.policyname)
    from pg_catalog.pg_policies p
    where p.schemaname = 'public'
      and p.permissive = 'PERMISSIVE'
      and p.roles && array['anon', 'public', 'authenticated']::name[]
      and (
        lower(btrim(coalesce(p.qual, ''))) in ('true', '(true)')
        or lower(btrim(coalesce(p.with_check, ''))) in ('true', '(true)')
      )
  ), '[]'::json),
  'public_buckets', coalesce((
    select json_agg(json_build_object('id', b.id, 'name', b.name) order by b.name)
    from storage.buckets b
    where b.public
  ), '[]'::json)
) as report;

The result is analysed in your browser. It is not sent to our server.

What this check does not cover

  • It checks only the page you enter and that page's own scripts, not routes behind login.
  • The deep check flags missing or wide-open rules; it does not judge whether your other policies match your app's logic.
  • It does not test Edge Functions, auth settings, API rate limits or anything outside the public schema and storage buckets.

Found something you are unsure how to fix? Tell us about your app and we will reply within 24 hours.