Scan your domain. Get the prompts to fix it in Claude Code.
SQUR's autonomous agent finds the vulnerabilities on your perimeter. The Fix Kit hands you the exact prompts to fix each one in Claude Code, Cursor, or Copilot. €25, one-time. No retainer. No subscription.
One payment. Yours to keep. One free re-scan included to verify your fixes.
How it works
Run the free 60-second scan on your domain. You get a summary of what is exposed.
The Fix Kit unlocks the full report plus one AI prompt per finding. Each prompt is written to paste straight into Claude Code, Cursor, or Copilot.
Paste the prompt. Review the diff. Apply. Then run your free re-scan to confirm the finding is gone.
This is what you get
Every finding comes with a prompt like this. Named file. Named line. A test baked in. Paste it into Claude Code and review the diff before anything changes.
# Fix: SQUR-0007 — JWT expiry claim not validated
**Severity:** High · **Location:** `src/api/auth/verify.ts:42` · **Class:** CWE-613 (Insufficient Session Expiration)
## Context
SQUR's agent issued a token, waited past its stated lifetime, and replayed it. The request
succeeded. `verifyToken()` decodes the JWT and checks the signature, but never checks the
`exp` claim. Expired tokens stay valid forever.
## Prompt for Claude Code / Cursor / Copilot
> In `src/api/auth/verify.ts`, the `verifyToken` function at line 38 to 50 verifies the JWT
> signature with `jwt.verify(token, secret)` but does not validate the `exp` claim, so expired
> tokens are still accepted.
>
> Refactor it so that:
> 1. `exp` is validated against the current time using the existing `jsonwebtoken` library
> (do not hand-roll the check — pass the verify options so `jsonwebtoken` enforces `exp`).
> 2. An expired token rejects with a 401 and the body `{ error: "token_expired" }`.
> 3. A token missing an `exp` claim is treated as invalid, not as never-expiring.
> 4. Existing valid-token behaviour and the current error handling are preserved.
>
> Add tests in `tests/api/auth/verify.test.ts` that assert:
> - A token issued with a 1-second lifetime is rejected after it expires (use fake timers).
> - A token with no `exp` claim is rejected with 401.
> - A fresh, valid token still passes.
>
> Show me the diff before writing any files.
## Why this prompt works
- Names the exact file and line range, so the tool does not guess.
- Tells the LLM to use the existing library instead of writing custom date math.
- Bakes the verification into the prompt as tests.
- Asks for the diff first, so you review before anything changes.
## After applying
Run your free re-scan (included with the Fix Kit) to confirm SQUR-0007 is resolved.
The re-scan link is at the bottom of your delivery email.
One Markdown file per finding. Numbered by severity. Highs first.
Inside the Fix Kit
Full scan report (PDF)
Every finding with severity, location, and compliance mapping for DORA, ISO 27001, NIS2, and SOC 2.
AI fix prompts
One prompt per finding, ready to paste into Claude Code, Cursor, or Copilot. Named files, named lines, a test in every prompt.
One free re-scan (30 days)
Apply the fixes, then re-scan the same domain to confirm the findings are resolved.
30 days of email support
Reply to your delivery email. A real engineer answers in 1 to 2 business days.
One price. Everything included.
- ✓ Full scan report (PDF)
- ✓ AI fix prompts for Claude Code, Cursor, and Copilot
- ✓ One free re-scan within 30 days
- ✓ 30 days of email support
Pay once. VAT shown at checkout.
7-day refund if the prompts do not apply to your codebase.
Built in the EU. Verified, not guessed.
EU-only data.
Your scan runs and stays in the EU. No data leaves the region. SQUR is an EU company, built for teams under DORA, GDPR, and NIS2.
Verified findings only.
SQUR's agent confirms each finding before it reaches your report. 87.5% KIT-validated. No noise. No false-positive padding.
No lock-in.
The Fix Kit is one payment. No subscription starts. No card kept on file beyond the purchase.
Questions
Which AI tools do the prompts work with?
Claude Code, Cursor, and GitHub Copilot. Any tool that takes a natural-language prompt against your codebase. The prompts name the file and line, so the tool does not have to guess.
Do I need to scan first?
Yes. The Fix Kit is the full report and prompts for a scan you have already run. Run the free scan first, then buy the Fix Kit for that domain.
What if a prompt does not apply cleanly?
Reply to your delivery email within 30 days. We tune the prompt for your codebase, or refund you within 7 days.
Is this a subscription?
No. €25, once. If you later want continuous weekly scanning, the €49/month tier is on our pricing page. It is optional.
Is my data safe?
Your scan runs and stays in the EU. Findings are verified before they reach you. We do not keep payment details beyond the purchase.
Get the findings and the fixes
€25. One payment. Full report, the prompts to fix every finding, and a free re-scan to prove it worked.
Get the Fix Kit · €25