Penetration testing for Sydney web applications, with a NSW buyer checklist
Clever Ops tests the websites, web applications and APIs that Sydney businesses run, remotely from Gippsland, Victoria.
If a NSW Government agency is one of your customers, the checklist on this page maps the questions it may put to you to the evidence a web application test can supply, and marks the questions a test cannot answer.
On this page
NSW buyer checklist for web application evidence
Find the row closest to the question your agency customer has asked, check the evidence column against what you actually hold, and write a name in the owner column before you answer. Rows cite the NSW Cyber Security Policy for 2026 to 2027 by requirement or section number and the Core& One core terms, version August 2025, by clause.
How to use itHide the instructions
They show what an agency may ask and why; they do not say that any rule applies to your business. General information, not legal advice.
13 rows
Question a NSW agency may put to you
Where the question comes from
When was the application we use last tested, by whom, and can we see the results?
- Where the question comes from
- Policy requirement 1.10.3: assessing third-party providers using assurance reports, audits, test results or other evaluations.
- Web application evidence that answers it
- The report's scope as tested (targets, roles, build and dates), its executive summary, and the findings table with CVSS v3.1 scores beside business-risk ratings.
- What that evidence cannot answer
- Anything outside that scope, releases after the tested build, and whether one test amounts to a testing program.
- Who on your side owns the answer
- Whoever commissioned the test, such as your technical lead.
Were the findings fixed, and how do you know?
- Where the question comes from
- Requirement 1.10.3, and the ongoing risk-based reviews of third-party compliance in section 8.4.2.
- Web application evidence that answers it
- The retest ledger, marking each finding fixed, partially fixed, still reproducible or not retested on an identified build.
- What that evidence cannot answer
- Fixes shipped after the retest. A finding marked not retested is unverified, whatever a developer reports.
- Who on your side owns the answer
- Your lead developer, and whoever approves releases.
How, and how quickly, will you tell us about a suspected security incident or data breach?
- Where the question comes from
- Requirement 1.10.2: a contractually supported process for providers to notify the agency. Core& One clause 10: notice within 48 hours of becoming aware of a Security Breach, or any shorter period required by law.
- Web application evidence that answers it
- None. A test report can note logging gaps the tester observed, but it does not show how you would notify anyone.
- What that evidence cannot answer
- Your notification process, who makes the call and how fast it runs.
- Who on your side owns the answer
- The owner of your incident response plan, with whoever signs the agency contract.
Describe the security program that covers the solution we buy from you.
- Where the question comes from
- Core& One clause 10: a formal security program in accordance with industry standards, comprehensive in covering the Solution.
- Web application evidence that answers it
- A dated test report and retest ledger, offered as one input to the program, with a methodology section naming its OWASP Web Security Testing Guide and ASVS 5.0 references.
- What that evidence cannot answer
- The program itself: policies, patching, access reviews, backups and monitoring. One test is not a program.
- Who on your side owns the answer
- Whoever is accountable for security in your business.
Can our staff be required to use multi-factor authentication to sign in to your application?
- Where the question comes from
- Requirement 3.5.2: agencies use MFA to verify users to third-party online services that process, store or communicate their sensitive data.
- Web application evidence that answers it
- What the login and session tests observed for each role tested: whether a second factor was asked for, and whether it could be skipped, for example by calling the API directly.
- What that evidence cannot answer
- Whether the agency classes its data in your application as sensitive, and settings in the agency's own identity provider if its staff sign in through it.
- Who on your side owns the answer
- Your product owner for sign-in. The agency's cyber security team decides whether the answer meets its needs.
Is our data encrypted between our staff and your application?
- Where the question comes from
- Requirement 3.12.3: encrypt all data communicated over network infrastructure, read with section 2.2 on services an agency relies on.
- Web application evidence that answers it
- Certificate and transport security results for each public hostname in scope.
- What that evidence cannot answer
- Traffic inside your hosting platform, and how stored data is encrypted. Those come from your hosting provider's documentation.
- Who on your side owns the answer
- Your developer, or whoever manages your hosting.
Can another of your customers see our records?
- Where the question comes from
- Section 9.4, advisory guidance: apply security controls based on data classification, and ensure data is properly separated.
- Web application evidence that answers it
- Findings and negative results from access tests between two customer organisations and between roles, recorded for each account and record tried.
- What that evidence cannot answer
- Features that were not tested, and separation inside the database that the application does not expose.
- Who on your side owns the answer
- Your lead developer.
Where is our data hosted and supported, and does any of it leave Australia?
- Where the question comes from
- Section 9.4: consider data sovereignty, including where data is hosted and supported. Core& One clause 9: a Seller with access to Personal Information must not transfer it outside Australia or access it from outside Australia.
- Web application evidence that answers it
- For your application, nothing in a test report. For the test itself, the tester's data handling terms: Clever Ops evidence is stored encrypted in Australia and processed outside Australia only where the statement of work approves it.
- What that evidence cannot answer
- Where your hosting, database, email and support providers keep and reach the data.
- Who on your side owns the answer
- Your technical lead, with your hosting and support contracts in hand.
What personal information will your tester see, and what happens to it afterwards?
- Where the question comes from
- Core& One clauses 7 to 9 on Confidential Information, Buyer Data and Personal Information. Section 8.2.6: agency privacy officers help identify and manage privacy impacts in new projects, systems and services.
- Web application evidence that answers it
- The environment named in the scope, evidence shown with secrets and personal details redacted, the list of test data left behind, and the deletion schedule in the data handling terms.
- What that evidence cannot answer
- Whether the agency's own privacy obligations are met. That is for its privacy officer.
- Who on your side owns the answer
- Your privacy contact, with the person who chose the test environment.
Who tested the application, and did they build or host it?
- Where the question comes from
- Supplier due diligence. No NSW document quoted on this page sets a rule on tester independence.
- Web application evidence that answers it
- The provider's written statement of whether it built, hosts or maintains the system. A Clever Ops report records who built and who hosts a system when that is Clever Ops, and calls the work a security review.
- What that evidence cannot answer
- Independence. A review by the builder or host helps with fixes but is not independent evidence.
- Who on your side owns the answer
- Whoever chose the provider.
Is your tester on the ICT Services Scheme, or CREST accredited?
- Where the question comes from
- The agency's own procurement wording. The ICT Services Scheme lists security testing as category K03.
- Web application evidence that answers it
- The provider's name and any scheme listing or accreditation it holds. Clever Ops is not a prequalified supplier on the scheme and is not CREST accredited, so we refer that work to a provider that is.
- What that evidence cannot answer
- Where the wording requires either, a report from a provider without it does not meet that wording.
- Who on your side owns the answer
- Whoever reads the procurement documents, before the test is booked.
Will you test again after a significant change?
- Where the question comes from
- Requirement 1.12.3, on the agency's own change management: test applicable cyber security controls and secure configurations upon finishing a significant change.
- Web application evidence that answers it
- The build identifier in each report and retest, which shows exactly which version the evidence covers.
- What that evidence cannot answer
- Changes released after that build.
- Who on your side owns the answer
- Your product owner, who plans releases and the testing calendar.
What is your Essential Eight maturity?
- Where the question comes from
- Requirements 3.3 to 3.10 of the policy, which incorporate the ACSC Essential Eight mitigation strategies.
- Web application evidence that answers it
- None. A web application test does not assess patching of staff devices, application control, macro settings or backups.
- What that evidence cannot answer
- All of it. Clever Ops does not assess Essential Eight maturity.
- Who on your side owns the answer
- Your IT provider or managed service provider.
What penetration testing in Sydney covers here
Clever Ops is a small Australian consultancy based in Gippsland, Victoria.
It has no Sydney office and no staff in New South Wales, and penetration testing for Sydney businesses is delivered remotely: the scoping call, the testing, the report and the retest. A web application test runs over the internet against the targets named in a signed authorisation, so it does not need a tester in your building.
The penetration testing services on offer to Sydney businesses cover what your customers reach through a browser or an API: marketing websites, WordPress sites and WooCommerce stores, custom web applications and their APIs, apps made with AI builders, and AI chatbots and agents.
A web application penetration test for a Sydney business is the same test, with the same report, as for a business anywhere else in Australia. A security fix sprint and retests sit beside it.
If you searched for cyber security in Sydney because you need someone to look after office computers, email, backups and the network, that is managed IT, and Clever Ops does not offer it. Nor does it test office networks, servers, Wi-Fi or Active Directory, run phishing campaigns or watch systems around the clock. A managed IT provider or a network testing specialist is the right first call for that work.
Vulnerability assessment and penetration testing are two depths of the same work. An assessment looks across the application for known kinds of weakness, confirms which ones are real and ranks them. A penetration test signs in as each role and tries to defeat the rules the application is meant to enforce, such as keeping each customer's records away from every other customer, and records the evidence of what worked. The scoping call decides which depth your application, and the customer asking about it, actually need.
Why NSW Government agencies ask their suppliers for test evidence
Cyber Security NSW issued the NSW Cyber Security Policy for 2026 to 2027 in July 2026.
Section 2.1 says it applies to all NSW Government departments, public service agencies, statutory authorities, and NSW Government organisations that submit an annual report to a secretary of a lead department or portfolio, directly to a minister or directly to the Premier. Section 2.3 says it is not mandatory for state-owned corporations, non-government organisations, local government and universities, which may adopt it as a standard where appropriate.
Its requirements are written for agencies, and this page reads them that way: as the reason an agency asks you questions.
Section 2.2 explains why a private supplier hears about it. The policy covers information and digital assets created and managed by the NSW public sector, including outsourced assets, and says agencies must comply with it across all systems, services and environments they own, operate or rely on, including third-party and shared services.
If agency staff sign in to your application, or agency data sits in it, your application may be one of the services an agency relies on.
Mandatory requirement 1.10 is to identify and manage third-party service provider risks. Its detailed requirements are an inventory of third-party service providers (1.10.1); a contractually supported process for providers to notify the agency of suspected or actual security incidents and data breaches (1.10.2); processes to monitor and assess providers' adherence to cyber security requirements, including using assurance reports, audits, test results or other evaluations (1.10.3); and contract clauses for cyber security requirements, with break clauses for failing to meet them (1.10.4).
Test results are one kind of evidence 1.10.3 names. The policy text names no minimum test, test frequency or tester credential for a supplier.
Section 8.4.2 says the duty to manage third-party cyber security risks covers, at a minimum, ICT service providers and other providers that process or store an agency's sensitive information. Agencies must set clear security requirements in contracts, monitor compliance, carry out ongoing risk-based reviews of third-party compliance, and set out how they will respond if a provider does not meet security requirements, with break clauses and offboarding given as examples.
The agency also works to its own clock. Requirement 2.3.2 has agencies report all cyber incidents through the Cyber Security NSW Cyber Portal within 24 hours of detection and classification. Keep that in mind when you answer a question about how, and how quickly, you would tell an agency about a suspected incident.
Privacy questions have an owner on the agency side too. Section 8.2.6 gives agency privacy officers the job of helping identify and manage privacy impacts in new projects, systems and services, and of coordinating the investigation of privacy incidents. The policy's useful links list the Privacy and Personal Information Protection Act 1998 and the Information and Privacy Commission NSW's guide to managing data breaches in accordance with that Act.
This page does not summarise what the Act asks of an agency; that is for the agency's privacy officer and lawyers. It is general information, not legal advice.
Contract terms that reach a security test
Agencies buy ICT goods and services under the ICT Purchasing Framework.
The buy.nsw page for the ICT Services Scheme says agencies must use the framework, under Procurement Board Direction PBD-2025-03, and that it comprises the Core& contracting framework for low-risk procurements up to a set value, and the MICTA and ICTA framework for high-risk procurements or those above that value.
If you already supply an agency, one of these may be your contract. The clauses below come from the Core& One core terms, version August 2025, which call the agency the Buyer and the supplier the Seller. If your agreement is on other terms, read your own.
- Clause 5: the Seller must have the Buyer's prior consent to subcontract any significant part of the Agreement, and is responsible for its subcontractors.
- Clauses 7 and 8: Buyer Data includes data the Seller stores, processes or generates on the Buyer's behalf in performing the Agreement. The Buyer owns or controls it from the time it is created, it is the Buyer's Confidential Information, and the Seller must comply with any restrictions on the location of Confidential Information set out in the Solution Requirements.
- Clause 9: where the Seller has access to Personal Information under the Agreement, it must not use or disclose it for any purpose other than performing the Agreement, transfer it outside Australia or access it from outside Australia. Privacy Laws are defined as the laws governing privacy as applicable in New South Wales and the Commonwealth.
- Clause 10: the Seller must maintain a formal security program in accordance with industry standards, comprehensive in covering the Solution, and must notify the Buyer within 48 hours of becoming aware of a Security Breach, or any shorter period required by law. As soon as reasonably possible after becoming aware of a Security Breach, it must conduct a root cause analysis and share the results and its remediation plan with the Buyer on request.
Where these clauses meet a test of your application
If testing production could put agency records in front of a tester, read clauses 5, 7, 8 and 9 with your solicitor before choosing production over a staging copy loaded only with seeded test data. The clauses say what each party must do. What they mean for a test you commission is a question for that solicitor.
A test report is one dated input to a security program, not the program itself, and its remediation plan is a different document from the root cause analysis clause 10 asks for after a Security Breach. Evidence from a Clever Ops test is stored encrypted in Australia, deleted on the schedule set in the data handling terms, and processed outside Australia only where your statement of work approves it.
Bring your contract's location restrictions to the scoping call so the data handling terms can be written to match them.
The ICT Services Scheme, panels and accreditation
An agency buying ICT services itself is pointed to the ICT Services Scheme. buy.nsw describes the scheme as a whole-of-government mandated arrangement and says NSW Government agencies must use it to buy ICT goods and services unless another mandated whole-of-government contract applies, under Procurement Board Direction PBD-2021-04. Its category K03, security testing, includes penetration testing, web security testing and secure code reviews.
Clever Ops is not a prequalified supplier on the ICT Services Scheme, is not on any government panel or supplier scheme, has no government endorsement, and holds no security certification or accreditation. Where a tender or contract requires a CREST-accredited provider, the same applies: Clever Ops is not CREST accredited, so we refer that work to a provider that is.
That makes the fit narrow, and it is better said early. If an agency wants its own systems tested, it buys through the scheme or another mandated whole-of-government contract, and Clever Ops is not the provider.
If you supply an agency and want your own application tested so you can answer its questions, read the agency's wording first. Where it asks for a scheme supplier or a CREST-accredited provider or names another credential, that wording decides, and we refer you.
Where it asks for recent test results and evidence that findings were fixed, a web application test can supply that evidence for the scope it covered.
Tester independence when an agency reads your report
An agency carrying out the ongoing risk-based reviews in section 8.4.2 can fairly ask who did the checking.
If the people who built or host the application also tested it, the report is a builder checking its own work, however capable the builder is.
Clever Ops builds and hosts websites and applications for clients. When it tests something it built or hosts, the work is called a security review, not an independent test, and the report records who built and who hosts the system.
If the agency or your contract asks for independent testing, choose a provider with no part in building or running the system, and say in your answer which kind of evidence you are offering.
Worked case: an invented Parramatta rostering company
An illustration written for this page, not a client or a real test.
An invented software company in Parramatta runs a shift-rostering web application and API at rosters.lab.test. Its customers include a NSW public service agency, whose staff sign in to build rosters, and a local council. The agency has sent a supplier security questionnaire with its contract renewal.
The company starts by sorting its customers. The agency is a public service agency, which section 2.1 of the policy names. The council is local government, for which section 2.3 says the policy is not mandatory, though it may adopt it. The company answers both from the same evidence, but it does not tell the council the policy binds it.
The questionnaire asks four things. When was the application last tested, and can the agency see the results? How would the company tell the agency about a suspected incident or data breach? Can agency staff be required to use multi-factor authentication? Does any agency data leave Australia?
The evidence is thin on the first question. The last test ran in August 2025, before the API the council uses was added, and one high-risk finding is still marked not retested. The answer says exactly that, gives the month a new test is planned, and describes its scope: the application and the API, with agency staff, agency managers, council staff and the company's own administrators as roles, and accounts in two customer organisations so access between the agency's rosters and the council's can be tested.
The other three questions go to different owners. Incident notification is answered from the company's incident response plan and the renewal contract, which is on the Core& One terms, so clause 10 is read beside it. Multi-factor authentication is a product question for the lead developer: the agency's interest traces to requirement 3.5.2, which has agencies use MFA to verify users to third-party online services that process, store or communicate their sensitive data, and the new test will record what the login tests observed for each role, including whether the second factor could be skipped by calling the API directly.
Data location is answered from the hosting contract and, for the test itself, from the tester's data handling terms.
Testing is planned on a staging copy loaded only with seeded data, so no agency records sit in front of the tester. Destructive actions are off by default for real people's accounts and records that existed before the test, and reset, lockout and deletion checks run only on test accounts and test records where the rules of engagement allow them.
Two things stay out of the answer pack. The company's marketing site was built and is hosted by a web agency, and a review by that agency would not be independent evidence, so the site is listed as out of scope rather than offered. The questionnaire's Essential Eight questions go to the company's IT provider, because a web application test does not assess those strategies.
Service-area notes: Sydney, Parramatta, North Sydney and Macquarie Park
These notes say how each area is served.
They make no claim of a local presence.
- Sydney CBD and inner Sydney: served remotely from Gippsland, Victoria. Clever Ops has no office, staff or meeting room in Sydney.
- Parramatta and Western Sydney: served remotely. Scoping, testing, the report and the retest follow the same process as for any other Australian business.
- North Sydney: served remotely. The test window is written into your scope with its time zone, so whoever watches your systems knows when test traffic is expected.
- Macquarie Park: served remotely. Site visits are not part of the service, and office networks and physical security are outside it.
What to gather before a scoping call from New South Wales
Bring the documents that set the question, so the scope answers your customer rather than a guess at what it wants.
- The agency's questionnaire or contract schedule, word for word, including any requirement that names a scheme, panel or credential.
- Which of your customers sit inside section 2.1 of the NSW Cyber Security Policy, and which, such as councils or universities, fall under section 2.3.
- The terms of each agency contract, whether Core& One, another Core& contract, MICTA, ICTA or something else, and any location restrictions for Confidential Information in the Solution Requirements.
- The roles agency staff hold in your application, and whether they sign in with multi-factor authentication or through their own identity provider.
- Whether a staging copy without agency records exists, or whether production is the only target.
- Who on your side owns incident notification, privacy questions and the contract, so every row of the checklist has a name beside it.
Frequently asked questions
No. Clever Ops is based in Gippsland, Victoria, and delivers penetration testing for Sydney businesses remotely. Testing runs over the internet against the targets in your signed authorisation, and the scoping call, report and retest work the same way for a Sydney business as for any other.
No. Clever Ops is not a prequalified supplier on the ICT Services Scheme and is not on any government panel or supplier scheme. If an agency, or your contract with one, asks for a scheme supplier or a CREST-accredited tester, we refer that work to a provider that meets the wording.
The policy's requirements are written for NSW Government agencies. Requirement 1.10.3 lists test results among the evidence an agency can use to assess a third-party service provider's adherence to cyber security requirements, and whether a particular report is enough is the agency's decision. This is general information, not legal advice.
Section 2.3 says the policy is not mandatory for state-owned corporations, non-government organisations, local government and universities, and that these may adopt it as a standard where appropriate. A council may still send you similar questions, so ask which document its questions come from before you answer.
No. The service is testing of websites, web applications, APIs and AI chatbots. Managed IT, monitoring, network penetration testing, phishing simulations and incident response are not offered, and a managed IT provider or a network testing specialist is the better first call for them.
Yes, remotely, for websites, web applications and APIs. An assessment looks across the application for known kinds of weakness, confirms which are real and ranks them. A penetration test goes deeper into the controls that matter most, such as whether one customer can reach another customer's records. If an agency is asking, its wording and the data your application holds decide which depth to buy.
Scope the right test for your application
Tell us what you have built, who uses it and when it needs to launch. You will get a plain-English recommendation: a website audit, a web application test or a referral to a specialist.