Manual testing first
Testers work through the in-scope functions by hand, using tools to support the work rather than replace it.

CREST-accredited penetration testing of web applications and APIs, focused on the flaws automated scanners miss: access control, business logic and data exposure.
Web applications and APIs hold customer data, take payments and run core processes. A launch, a major release or a new integration can introduce weaknesses that only show up when someone tries to use the application in ways it was never meant to be used.
Our testers work through the application manually, guided by recognised methods such as the OWASP Testing Guide and the OWASP Top 10. Automated tools support the work, but the focus is on the issues they cannot find: one user reaching another user’s data, a step in a process that can be skipped, or a role that can do more than it should.
GIFGIF is CREST accredited for penetration testing and certified to ISO/IEC 27001. Findings come with clear reproduction steps and remediation guidance your developers can act on, and a summary that explains the business significance to product owners and leadership.
No permission. No test.
Authorisation. Testing requires written authority, agreed environments, test accounts and rules of engagement. Results are point-in-time observations, not a guarantee of security or coverage of untested functions. Remediation remains with the application owner; any retesting must be separately confirmed.
Testing that follows how the application is really used. Automated scanners find some issues. Many of the most serious application flaws sit in access control and business logic, which need a tester who understands what the application is meant to do.
Testers work through the in-scope functions by hand, using tools to support the work rather than replace it.
Each agreed user role is tested against the data and functions it should and should not be able to reach.
Reproduction steps and remediation guidance are written for the people who will fix the issue.
The work is defined by the systems and applications agreed for this assessment. These descriptions do not expand that scope.
Test whether each user can do exactly what they should, and nothing more.
Testing covers login and password reset flows, multi-factor authentication, session handling and the separation between user roles. Broken access control is one of the most common serious findings in web applications, so we test each in-scope role against the functions and data it should and should not reach.
Where the application supports several organisations or tenants, we test the boundaries between them. Test accounts for each relevant role are agreed in advance.
Examine how the application behaves when a process is used out of order or with unexpected input.
Payment flows, approvals, discounts, quotas and multi-step processes can often be manipulated in ways a scanner will never spot. We test the logic of the in-scope processes as well as classic input issues such as injection and cross-site scripting.
Each finding is explained in terms of what an attacker could achieve in this application, not just the generic vulnerability class.
Assess the interfaces that mobile apps, partners and front ends rely on.
Many applications expose more through their APIs than through the user interface. Where APIs are in scope, we test authentication, authorisation on each endpoint, data returned in responses and rate limiting, working from documentation such as an OpenAPI specification where available.
Third-party services the application depends on are only tested where their owner has given permission. Otherwise they are noted as dependencies outside scope.

Written authorisation, identified targets and agreed rules of engagement are prerequisites. Activity stays within that scope.
From the brief to useful security findings. We agree the objective, written authority and rules of engagement before assessing the named systems and explaining the results.
We clarify the decision, parties or systems in scope, intended use and lawful purpose of the work.
Written permission, identified targets, exclusions and rules of engagement are established before testing.
Information is considered for source, date, relevance and corroboration; uncertainty is not converted into a conclusion.
The account explains what was observed, what it may mean, what remains unresolved and the limits of the work.
When web application testing is commonly commissioned. Examples to help frame an enquiry; these are not case histories.
A product team may want independent testing before customers or partners start using a new application or feature.
An enterprise customer, insurer or regulator may ask for a recent CREST penetration test report on the application they rely on.
A business opening an API to partners may want its authentication and data exposure tested before it goes live.