Application & Mobile Security Review

An application can pass a network test and still hand over every customer record, because the flaw is not in the infrastructure. It is in the logic.

Changing one identifier in a URL and seeing somebody else’s invoice is not a misconfiguration a scanner reports. It is a design decision nobody questioned. We test what your application lets people do that it should not, across web, mobile and the code behind them.

Application source code under review on a developer console
Findings your developers can reproduce, not a PDF they argue with.
When you need this

Four situations that bring teams to us

You are shipping something new

Or you have shipped a great deal very quickly, and nobody outside the team has reviewed it.

A customer sent a security questionnaire

They want evidence of independent application testing, and a deal is waiting on your answer.

Your application handles money, health or personal data

The consequences of a logic flaw are not theoretical, and neither is your regulator.

Strong developers, no application security specialist

Good engineers miss authorisation flaws in their own code for the same reason writers miss their own typos.

What we test

Coverage across the application stack

Web applications

Authentication and session handling, access control both horizontal and vertical, injection, file handling, and business-logic abuse, where a workflow can be completed in an order nobody designed for.

Mobile, Android and iOS

Local data storage, transport security and certificate handling, communication between components, resistance to reverse engineering, and the backend the app talks to, which is usually where the significant findings are.

Databases at application level

Database testing through the application layer: what the application will return, to whom, and under what conditions.

Secure development and code review

Reviewing the code behind the behaviour we observe, which finds classes of defect that black-box testing cannot reach in a sensible timeframe.

Interfaces and APIs

The endpoints behind your web and mobile clients: authorisation checked per object and per action, rate limiting, and the excess data an endpoint returns that the interface simply never displays.

How we work

Grey-box by default

You give us accounts at each privilege level, and we test what each one can reach. It mirrors the realistic threat, which is a legitimate user, a leaked credential or a malicious customer, and it spends your budget on depth rather than on us guessing your URL structure.

Testing follows established international application security methodologies, chosen for the application in front of us rather than applied uniformly. We confirm the approach with you during scoping.

What we ask you for

One account per role, a stable environment, and someone on your side who can answer a question about intended behaviour. That is usually the whole list.

Deliverables

What your developers actually get

  • Findings in a form your team can work from directly: title, severity, affected endpoint, reproduction steps and evidence.
  • Severity that accounts for context. What the flaw exposes, for which users, and what an attacker needs in order to exploit it.
  • Guidance on the fix, and on the pattern that stops the same class of defect returning next sprint.
  • A debrief with the people who will do the fixing, not only with whoever commissioned the test.
  • Retesting after remediation, on terms agreed for the engagement.
Scoping

How we scope an application test

Application engagements vary more than infrastructure ones. The number of user roles, the number of workflows, whether an API and a mobile client are included, and whether we have source access all change the work substantially. We scope on a call and quote a fixed price for that scope.

We will also tell you when not to test

Testing a build that is about to be replaced is an expensive way to generate a document. If your application is not ready, we would rather say so before you commit than bill you for finding out.

FAQ

Common questions

Do you need our source code?
No, but it improves the result. Source-assisted testing finds authorisation logic errors and injection paths that are impractical to reach from the outside within the time available.
Should we give you staging or production?
Staging, if it genuinely mirrors production: same code, same configuration, representative data. Where it does not, we will say what that leaves untested rather than quietly ignoring it.
What happens to our data?
Handling, storage, retention and destruction of test data and evidence are agreed with you in writing before testing begins, as part of the engagement terms.
Can you test on every release?
Cadence is agreed case by case. For teams that ship frequently we can structure an engagement around releases rather than a single annual assessment.
We rely on third-party components. Are those covered?
We flag vulnerable and unmaintained dependencies as part of the review. Testing a third-party service itself requires that vendor’s permission, and we will tell you where that line falls for your stack.
Is this the same as a penetration test?
It is penetration testing aimed at the application layer rather than the network. Most organisations need both, and they are often scoped together. See penetration testing.
Credentials

Certifications held across our team

CISSPCertified Information Systems Security Professional · (ISC)²
CRISCCertified in Risk and Information Systems Control · ISACA
CISACertified Information Systems Auditor · ISACA
CDPSECertified Data Privacy Solutions Engineer · ISACA
GDSAGIAC Defensible Security Architect
GCLDGIAC Cloud Security Essentials
OSCPOffensive Security Certified Professional · OffSec
ISO 27001Lead Auditor

Describe the application

Tell us what it does and who uses it, and we will tell you what it needs tested and how we would scope it.

Contact Cyberange