Made for apps built with Lovable
A security scanner for the URL you published
Sekrd scans the deployed URL. It does not need a Lovable login. A free scan reads the page and its JavaScript. A deep scan adds read-only probes, including whether the Supabase publishable key already in the page can read common tables. Lovable's own Quick scan and Deep scan stay in the project. This page is the outside check.
Free scan. Pre-Launch Audit is $39 once. Continuous Pro is $24/month billed annually ($288/year), or $29/month billed monthly.
Risks Sekrd checks on a Lovable app
Lovable apps usually ship a Supabase publishable key in the client. That is expected. The failures below are the ones the scanner is built to report. The comparison with Lovable's own scan is on Sekrd vs Lovable security scan.
Free scan
Secret Supabase key in the shipped JavaScript
The secrets check flags a service_role JWT and an sb_secret_ key in the HTML or JavaScript the browser downloads. It does not flag sb_publishable_ or the anon JWT. Those are public. A secret in the bundle bypasses row level security for anyone who views source.
Deep scan
Tables the publishable key can read
If the page contains a https://<project>.supabase.co URL and an anon JWT, the deep scan sends a read (limit 1) to a fixed list of common table names on /rest/v1 using that key. HTTP 200 with row data is a critical finding. It does not write.
Deep scan, optional database password
RLS off, or a policy whose expression is true
Policy SQL is read only when you attach a Supabase integration and a database password. That path queries pg_policies and flags tables with row security disabled and policies whose expression is true, including USING (true). Without the password, Sekrd does not claim to have read the policy text. A default project does not expose pg_catalog through PostgREST.
Free scan
A client env prefix on a server secret
NEXT_PUBLIC_, VITE_, EXPO_PUBLIC_, NUXT_PUBLIC_, and PUBLIC_ inline the value into the browser. The free scan reports that prefix when the variable name is a server secret: model API keys, a Stripe secret, service role, a database URL, a webhook secret, and the other names in the jssecurity rule. Public-by-design names such as ANON and PUBLISHABLE are not in that list.
Free scan
Payment webhook that accepts an unsigned POST
The payments check POSTs an unsigned body to a fixed list of Stripe, Paddle, and Lemon Squeezy paths. HTTP 200 with a non-HTML body, distinct from a catch-all, is critical. An HTML single-page-app shell is ignored.
Deep scan
MCP tools/list with no credentials
If the origin exposes an MCP endpoint, the deep scan calls tools/list without credentials. A successful list is high, or critical when a tool name looks like SQL, shell, file, or admin. initialize alone is not a finding. The check does not call any tool. MCP checklist.
Free scan
Headers, storage, and the bundle
Missing or weak Content-Security-Policy, HSTS, X-Frame-Options, nosniff, Referrer-Policy, and Permissions-Policy. Auth tokens written to localStorage. postMessage listeners without an origin check. eval, innerHTML, and document.write are checked and then dropped on a heavily minified bundle instead of reported as noise. An exact pin of react-server-dom-webpack, parcel, or turbopack at 19.0.0, 19.1.0, 19.1.1, or 19.2.0 is reported (CVE-2025-55182). Patched revisions 19.0.1, 19.1.2, and 19.2.1 do not match.
Free scan
System prompt that contains a secret or an authorization rule
A systemPrompt, system_prompt, or systemMessage assignment in the bundle is reported when the prompt text contains credential-shaped material or tells the model to bypass auth or RLS. A prompt that merely mentions the word password is not enough.
What a scan does not do
A free scan reads the public URL: headers, HTML, and JavaScript. Supabase policy SQL, Firebase rules, TLS and DNS email checks, network ports, and the read-only active probes run on a deep scan. Deep scan is included with the $39 Pre-Launch Audit and with Continuous Pro. See pricing.
Active probes are reads and unsigned requests against known paths. Sekrd does not guarantee the app is safe, and a clean result is not a pentest. Findings the scanner could not observe are not listed.
Related reading
- How to secure a Lovable app
Keys, RLS, storage, and Edge Functions before the first public URL.
- What Lovable's security scan checks
Quick scan and Deep scan, from Lovable's docs.
- CVE-2025-48757
Missing or permissive RLS on generated apps.
- Supabase RLS checklist
Policies that mention the caller, not USING (true).
- Lovable audit landing
The $39 pre-launch audit, including the compliance package.
Frequently asked questions
Does Sekrd need access to my Lovable project?
No. Paste the deployed URL. A deep scan can optionally take a Supabase database password you provide if you want policy SQL read. Sekrd does not ask for a Lovable login.
Is the Supabase anon key a finding?
No. The publishable key and the legacy anon JWT are public by design. Sekrd reports a service_role JWT or an sb_secret_ key in the client, and, on a deep scan, tables that return rows to the publishable key.
Does the free scan read Row Level Security policies?
No. The free scan can see a secret key in the bundle. Reading pg_policies requires a deep scan with a database password you provide. The deep scan can also try common tables with the anon key found in the page, without that password.
How much does it cost?
The URL scan is free within the published quota: 3 anonymous scans a day, or 5 a day and 90 a month when signed in. Pre-Launch Audit is $39 once. Continuous Pro is $24/month billed annually ($288/year) or $29/month billed monthly.