I loved reading the blog post Threat Hunting: Taking the First Step to Restoring Cyber Hygiene After a Pandemic, written by Craig Robinson, Program Director, Security Services at IDC. The three main points that Craig made in his excellent writeup are:
I completely agree these actions are the right pieces, but I tend to differ regarding the right split around what skills and expertise customers should be measuring. In my experience, it’s important that organizations don’t just check to make sure that threat hunters are doing the three things already mentioned… they should also be using research and background knowledge to make a guess about something that has NOT yet happened.
So, in the spirit of building upon the great recommendations that Craig put forth, I present my 3-part plan for becoming a lean, mean proactive threat hunting machine.
Consider putting into place automated ways to search your data stores for threats that are:
This approach combine’s Craig’s first and third recommendations into a hybrid model that foundationally is based on known TTPs or known and identified IOCs.
This inference and decision tree-based model relies on a fundamental continuous monitoring principle, which I want to highlight but was missed in Craig’s focused analysis. This is what makes threat hunting a fitness exercise. The shared principle is that either the TTP or the IOC is either a known quantity, or was reverse malware engineered and searched for again during escalation.
Think of it as doing bicep curls with a specific weight on a continuous basis – you know you’ll gain muscle mass on your arms. Good for that area of the body, but you also need to work out your legs to stay balanced.
Security orchestration – or the process of integrating and automating all of your disparate security tools to create a stronger security posture – can help in a variety of ways. But it’s important to use it wisely.
Orchestration speeds up threat hunting by correlating threat intelligence, the alert data that you receive across all your security tools, and by taking action in some cases. Still, it’s really a decision tree of dominos, not proactive threat hunting. Orchestration alone is not enough for threat hunting but is a key component to agility and speed during investigations or within an incident response escalation plan. Just as cardio gets blended with strength training in any good fitness plan, your security orchestration needs to be complemented by a human component, or a fitness coach.
Predictive, or, proactive threat hunting is a point in time assessment. It can potentially be cadential in nature, but it’s generally governed with a start and end.
To make your proactive threat hunting truly powerful, think of it as circuit training: making sure it’s done periodically, even if there are no symptoms or signs of trouble. Use a timed cadence, set to regularly repeat on a schedule, help intercept threats or vectors and to identify unknown unknowns.
Identification of malware during a targeted threat hunt to “RED” assets as defined by Craig may dictate a different intervention than a “GREEN” asset in terms of remediation timelines, but both are important to identify, query, and track, and I recommend a balance between the two. Test both, train for both.
Proactive Threat Hunting taught me to always run that last lap. Don’t forget – it’s always ok to ask for help. Our personal records come from this push and define our forward outlook on the next goal.
When you partner with Trustwave SpiderLabs for a proactive threat hunt, our expert hunters will employ our proprietary Threat Hunting Platform, combined with industry-leading EDR and powerful SpiderLabs cyber threat intelligence to find threats and weaknesses within your network infrastructure.