Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
The rules of engagement are the written agreement that defines what a penetration test may touch, when, by what means, and who to call when something goes wrong.

What Are the Rules of Engagement in a Penetration Test?
The rules of engagement are the document that makes a penetration test lawful, safe, and useful. They record what is in scope, what is explicitly excluded, when testing may happen, which techniques are permitted, how findings and data will be handled, and who both sides call if something breaks. Signed before any testing starts, they are the reference point whenever a question arises mid-engagement.
Skipping this step is how engagements go wrong. A tester takes down a production service nobody mentioned was fragile, probes an address that belongs to the client’s hosting provider rather than the client, or triggers an incident response at three in the morning because nobody told the on-call team. Every one of those outcomes is a scoping failure, not a testing failure.
Scope often includes systems the client does not own. Software-as-a-service platforms, managed hosting, payment providers, and outsourced application support all have their own terms about security testing, and the client’s signature does not bind them.
The major cloud providers have simplified part of this: AWS, Microsoft Azure, and Google Cloud all permit customers to test their own deployed services without prior notification, within published limits that exclude denial of service and attacks on shared infrastructure. Smaller providers rarely have a published policy, so written permission must be obtained before an address in their range goes on the list. The same applies in reverse: if your supplier commissions a test that touches your systems, you should be asked.
Testing stops when continuing would cause harm or confusion. Typical triggers are evidence of a pre-existing compromise, discovery of unexpected personal data, a production outage, a safety-critical system appearing in scope, or a law enforcement or regulatory matter. The document should say who is called, how quickly, and what happens to collected evidence in each case.
Deconfliction matters whenever a test might be mistaken for a real attack. For covert exercises, a small number of trusted client staff hold a list of tester source addresses and activity timestamps, so when the SOC escalates, somebody can confirm within minutes whether it is the engagement. Without that arrangement, organisations have invoked full incident response, taken systems offline, and notified regulators over authorised testing.
Some UK sectors formalise all of this. The NCSC CHECK scheme sets requirements for testing public sector systems, including qualified team leaders and a defined reporting format. CREST member companies work to published methodology and conduct standards. For financial services, the Bank of England’s CBEST framework, and TIBER-EU more broadly, require intelligence-led testing with the regulator, the firm, the threat intelligence provider, and the red team all operating under a jointly agreed control group and engagement plan.
PCI DSS expects a documented testing methodology covering scope, segmentation validation, and retention of results, so the rules of engagement feed directly into the evidence pack.
Write the document with the people who run the systems, not only with procurement. Ask what is fragile, what is monitored, and what a bad day looks like operationally. Agree in advance how aggressive exploitation may be on production, because the honest answer varies by system. Keep the signed copy accessible to the testers throughout, particularly for physical penetration testing, where it may need to be produced to a security guard.
At Privia Security, every engagement under our Penetration Test Services starts with this document agreed and signed, because it protects the client and the tester equally.
Someone with genuine authority over the assets in scope, usually a senior IT or security leader, plus the testing provider. Where systems are hosted or operated by a third party, that party’s permission is needed as well. A signature from someone who cannot authorise access to the systems listed offers no protection to either side.
Yes, and it often does when discovery reveals assets nobody knew about. Changes must be agreed in writing by the same people who signed the original document, even if that is an email confirming a new address range. Testers should never add targets on the basis of a verbal instruction, however senior the person giving it.
Only deliberately, and usually not against production. Most engagements exclude it, because the risk of outage outweighs the value of confirming what everyone already assumes. Where resilience genuinely needs testing, it belongs in a dedicated exercise against a representative environment, with the hosting provider informed and a clear rollback plan in place.
It depends on the objective. For a standard penetration test, telling them avoids wasted incident response effort. For a red team exercise, the whole point is to measure detection, so only a small control group knows. Either way, a deconfliction contact must be able to confirm within minutes whether suspicious activity is the engagement.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.