ProductsBug hunting

The bugs scanners miss,found by people.

Our specialists work through your live applications looking for flaws in access control, business logic and workflows, the kind that only a person notices. A bug is reported once it is reproduced, with proof and the steps to fix it.

GET /api/orders/10482signed in as Ana
HTTP/1.1 200 OK
Order
10482
Customer
Mihai Pop
Address
Str. Exemplu 4, Iași
Card
•••• 8812
Not Ana's order
BUG-0142 · High 7.1 Any customer can read another customer's order. Reproduced, with the fix.
Sample data. Names, addresses and values are fictional.

What you pay for

Bug hunting is priced per asset, like scanning: one domain or one subdomain. A domain does not include its subdomains, so each subdomain is an extra asset.

What you buy is a specialist's time on that asset. A single hunt finds what is there today. Monitored hunting keeps specialists on it every month, so every release gets looked at by a person, not only by a scanner.

  1. northwind.examplethe domain
  2. shop.northwind.examplea subdomain, counted on its own
  3. app.northwind.exampleanother subdomain
Assets to scan3
Sample data. Names, addresses and values are fictional.
One hunt
315 €per asset
350 €Winter Sales · −10%

One round of hunting. Right before a launch or after a big change.

  • One hunting round on the asset
  • Every bug reproduced before it is reported
  • Proof, severity, CVSS and fix steps per bug
  • Anything you release after the hunt
Book a hunt
Monitored bug hunting
855 €per asset / month
950 €Winter Sales · −10%

Specialists stay on the asset every month and look at what you ship.

  • Hunting every month, including new releases
  • Confirmed bugs reported as they are found
  • Proof, severity, CVSS and fix steps per bug
  • A retest on production after each fix
Start monitored hunting

Prices per asset, before VAT. Monitored hunting is sold for 1 or 3 years, paid up front. Testing starts only after written authorization.

Where people find what tools don't

Broken access control
Reading another customer's order by changing a number in the URL. Every request looks valid to a scanner.
Business logic
Stacked discounts, negative quantities, a payment step that can be skipped.
Accounts and sessions
Password reset flows, tokens that never expire, sessions that survive a logout.
Data the screen hides
APIs that return more than the app displays: emails, addresses, internal IDs.

How it runs

  1. 1ScopeYou authorize, in writing, which applications and accounts are in scope.
  2. 2HuntSpecialists use the app like a customer would, then like an attacker would.
  3. 3ReproduceNothing is reported until it is reproduced and its impact is clear.
  4. 4Report and retestEach bug arrives with proof and fix steps; after you fix, we test it again on production.

What a confirmed bug comes with

  • The exact request and response that show the bug
  • What an attacker could do with it
  • Severity, CVSS score and CWE class
  • Fix steps, and a retest once you ship the fix

Tell us what you run. We will scope it with you.

Talk to an expert