By Kim Crawley
Not everything that’s called “pentesting” is pentesting. There’s an abundance of different types of security testing and tools that use different methodologies for different stakeholders with differing agendas. Security testing, which includes pentesting and also vulnerability assessment, compliance auditing and other formats, is even broader. We’ll break down the differences between types of pentesting and strategies that are labeled pentesting but are fundamentally different.
First, what are you testing for?
Are you trying to penetrate a network or computer system like a cyber threat actor, but with permission from the owner for the purposes of discovering security vulnerabilities? Then chances are what you’re doing is pentesting. If you’re using a checklist of security standards of some sort and looking for vulnerabilities without simulating cyber attacks, that’s a vulnerability assessment. It sounds obvious, but some entities try to sell vulnerability assessments by incorrectly calling them pentests. Pentests aren’t “better” than vulnerability assessments–they’re different types of security testing. Each can be the best solution for different problems.
The Flavors of Pentesting
Pentesting is having specially trained people simulate cyber attacks. They can use applications, scripts and even conduct analog activities such as social engineering and physical security pentesting. Its strength and weakness is the people doing the testing and the platform they work on. Without good testers on an efficient platform, the test may not leave the buyer with confidence. Traditional pentesting relies on only the skills of a few people and outputs a readable report, not data. Synack was founded to get the best testers on the best platform for the best pentest possible. A pentest’s output – at least Synack style – is real-time access to findings, remediation information, analytics about testing and more.
Different types of pentesting can be categorized according to which facet of a computer system is being tested. The majors are network pentesting, application pentesting, social engineering pentesting that finds vulnerabilities in people and physical pentesting that finds vulnerabilities in buildings, doors, windows, rooms and the like.
Pentesting is also categorized according to the information available to the testers. Blackbox testing is done with little to no knowledge of a target from the perspective of an external attacker. Whitebox testing is done with in-depth target knowledge from the perspective of an internal attacker in the target’s IT department. And Greybox testing is in the middle from the perspective of a nontechnical insider.
There are also other ways to prepare for cyber threats that are different from pentesting. Let’s explore some of them.
Methodologies for Security Testing (That Aren’t Pentesting)
Breach and Attack Simulation (BAS) based on attack replay or scripting is a relatively recent development in security testing tech. Scripts that simulate specific exploits can be executed whenever an administrator needs to test a particular attack. This way, teams are better trained to know how to spot attack patterns and unusual log activity. When the cybersecurity community discovers new exploits, scripts can be used to simulate those exploits. Note that that takes time, so BAS may not be as current as adversarial tradecraft. The testing-like output is confirmation how many known vulnerabilities with easily scriptable exploits exist in your environment.
BAS is best suited for testing security responses to ensure teams know how to spot attack patterns and strange attacks in their log systems. This is a great training tool for blue teams but will not result in the discovery of unknown vulnerabilities in general. This shouldn’t be viewed as a pen test replacement and usually the scripted models lag the current adversary tradecraft.
Bug Bounty welcomes members of the general public under well defined policies to security test your software themselves and submit bug reports to your company according to the principles of responsible disclosure. If a bug can be proven and fits your company’s criteria of a prioritized vulnerability, the bug hunter could be awarded a monetary prize of anywhere from $50 to $100,500, but typical bug bounty rewards are about $200 to $1,000. The amount of money awarded for a valuable bug report is affected by several factors including the size of the company’s budget and user base and the criticality of the bug.
Dynamic Application Security Testing (DAST) is an automated technique, but it’s exclusively for testing working applications. So it’s often a tool used by application developers. DAST is used most often for web applications, but other internet-connected applications can be tested this way too. The targeted application must be running, such as a web application on the internet. The exploits that are executed are dynamic, so they may alter course depending on the progress of penetration.
Risk assessments are sometimes called threat evaluations. In a risk assessment, your security team collaborates with what they know about your organization’s data assets and how those assets could be threatened, both by cyber attack and by non-malicious threats such as natural disasters and accidents. Risks are identified, estimated and prioritized according to their probability of occurring and the amount of harm that could result.
Static Application Security Testing (SAST) has the same goals as DAST, but for application code before being compiled, not for applications that are running in production mode. If a vulnerability is clear from source code – and not all are – it can be detected by SAST.
Tabletop exercises are mainly for incident response teams, a defensive security function. They can be a fun challenge when done well, and help your incident response group face cyber threats with greater confidence. Specific attacks are proposed in the exercise, and the team needs to figure out how they should prevent, mitigate, or contain the cyber threat. If Capture The Flag is the main educational game for the red team, tabletop is the main educational game for the blue team. The output is a more confident and prepared team. Sometimes, refinements for an organization’s threat modeling also emerge. But actual vulnerabilities will not often be found during these exercises.
These and other newer technologies (artificial intelligence and machine learning in particular) are useful tools for security leaders. Computer software acts faster and doesn’t get tired, but the most flexible thinking comes directly from human beings.
Computer scientists know that computers can only simulate randomness, it takes a living being to actually be random. And human pentesters, like the Synack Red Team, are the best at simulating human cyber attackers and the serious exploits they regularly find.
For a deeper look at the Synack Red Team and its diverse skill set, read our latest white paper, “Solving the Cyber Talent Gap with Diverse Expertise.”