Missing a country or a custom check? Tell us what you need.Book a 15-min call

Our Independent Pentest: Fixed and Retested in One Day

Phorsys ran a black-box penetration test on secuveri. We received the report, fixed every finding and retested it ourselves, all on the same day.

4 min readNovember 2026

By secuveri teamPublished

secuveri and Phorsys logos above a checklist showing the black-box test done, every finding fixed and the fixes retested internally

Report, fixes and retest on the same day

On 9 November 2026 we received the final report from an independent penetration test of our platform. By the end of that day we had fixed every finding, deployed the fixes to production and written our internal retest report.

This post explains how the test worked, what we changed and why we asked for it. Our security page covers the controls we run every day.

Why we asked for an outside test

Our customers send us passports, ID cards and selfies. They trust us to keep that data safe. Saying a platform is secure is easy. Showing it takes evidence that does not depend on our own word.

So we asked an independent team to attack us. Phorsys, a French offensive security firm that specialises in penetration testing and fraud, ran the test.

"We handle identity documents and biometrics, so we want to know where we are weak before anyone else does. The Phorsys team attacked us like a real adversary would, and their report made every fix clear. That is why we could close everything in a day."

Artus Guenes, CEO of secuveri

How Phorsys tested

Phorsys ran a black-box test. Their team started from our public web address and nothing else: no code, no internal documents and no special access.

  • Scope. Our web application and its API, in production.
  • Timing. Scoping on 4 November, testing from 4 to 8 November 2026, outside business hours.
  • Rules. No denial of service, no real payments, no mass emails to real people and no access to other customers' data.
  • Method. Recognised frameworks: the OWASP Web Security Testing Guide for coverage, PTES for the structure of the engagement and CVSS for scoring.

They mapped what we expose, scanned it with automated tools, then tested by hand. That covered sign-up and login, sessions, access control between organisations, input handling, business logic and configuration. On 8 November they gave us a hot debrief, and the written report followed.

Timeline from scoping on 4 November 2026 to the report, fixes and internal retest on 9 November

What the test showed

The platform held. Phorsys did not compromise it in the time available, and none of the outcomes we feared most happened. Nobody reached data without an account, read another customer's data or took over an admin account or a server.

The report names effective access control, sound authentication on the features tested and robust, proven components as strengths.

The test also surfaced a set of issues around API hardening, authentication and abuse protection. Each came with a clear explanation and a concrete fix. That is exactly what we wanted from the test: specific things to improve, found by someone who thinks like an attacker.

What we hardened

We fixed every finding on the day we received the report and deployed the changes to production. In broad terms, we:

  • reduced what our API shows to callers who are not signed in,
  • tightened the rules around accounts and sign-in, enforced by the server,
  • added stricter limits and input checks on actions users can trigger,
  • applied stricter network rules to the services that support live checks such as liveness detection.

We keep the technical details private. Publishing them would help attackers more than it would help customers.

What we hardened after the test: API, authentication, abuse protection and infrastructure, each fixed and retested

How we verified the fixes

The retest was done by secuveri, internally, not by Phorsys. On the same day, we repeated each of Phorsys's tests against production, from the outside, with the same kind of tools. Every test that had worked before now failed. Our retest report records the method and evidence for each item, and we added one extra hardening step while we were at it.

Phorsys recommends an independent counter-audit after fixes, and we agree. That counter-audit has not happened yet. Until it does, the fixes are verified by our own testing.

A penetration test also has limits, and the Phorsys report states them openly. It covers a defined scope over a fixed time, and the result is a snapshot. A black-box test can miss things that someone with inside knowledge would see. That is why we treat testing as something we repeat, not a box we tick once.

What this means if you are evaluating secuveri

If you are a compliance officer or CTO assessing us, here is the short version:

  • An independent team attacked our production platform from the outside.
  • It did not get in. Access control and authentication held.
  • We fixed every finding and retested the fixes the same day the report arrived.
  • An independent counter-audit of the fixes is the next step.

Customers and auditors can ask us for the Phorsys report and our retest report. The same thinking runs through the rest of the platform: verification data is processed and stored in the EU (Vienna, Austria) with encrypted backups in Germany, our models are built in-house and every outcome is a signed result. Our compliance page has more detail.

Thank you to the Phorsys team for a careful, well-run test. If you need a black-box penetration test, or want to know how your KYC flow holds up against forged documents and deepfakes, talk to Phorsys.

This article is general information, not legal or compliance advice. Your obligations depend on your sector, licence and markets.

secuveri

See how secuveri protects your verification data.

Keep exploring

Contact

Talk to us

Tell us what you’re trying to verify. We’ll show you how secuveri can fit your workflow.

We reply within one business day and use your details only for your inquiry. See our privacy policy.