Awesome GPTs for Cybersecurity are ...
well, awesome
It’s only been 10 days since OpenAI announced GPTs - which are custom versions of ChatGPT that users can tailor for specific tasks or topics by combining instructions, knowledge, and capabilities. They require zero coding skills to create. And, as with all things related to generative AI, they have taken off at wildfire pace.
I’ve played around a little bit creating a few of my own, but I’ve found it much more fun (and useful!) to see the very cool GPTs that others are creating. Yesterday I was introduced to what might become a bit of a goldmine for cybersecurity GPTs, courtesy of: Thomas Roccia - Sr. Threat Researcher at Microsoft, Malware Warlock, Threat Intel, Python🧡- Former McAfee_labs, Goon Defcon, Creator of http://unprotect.it. I subscribe to his weekly visualized threat reports newsletter - The Intel Brief. In yesterday’s edition he mentioned this new treasure trove list of cybersecurity GPTs that he has created:
Awesome GPTs (Agents) for Cybersecurity
This morning I’ve had a chance to start trying out some of the cybersecurity GPTs on his list, and I have already pinned a few (more on that below) and I’m maybe twenty percent of the way through the list. I’m sure I will be coming back to explore this list often. For now, I want to just share a few of the cybersecurity GPTs that look useful and fun to me after my first look at it today.
Hacker Art: “Generate badass hacker art and profile pics”. I used this one for the image at the top of this post.
SOC Copilot: “Cybersecurity expert with keyword-based guidance.” I tried this one out by asking for indicators of compromise (IOCs) for the Scattered Spider threat actor group.
Threat Intel Bot: “A specialized GPT for the latest APT threat intelligence.” APT is the acronym for advanced persistent threat, generally attached to nation-state level threat groups or similarly well resourced adversary groups. I used one of the default prompts with this one, which asks to see the latest MITRE ATT&CK tactics and techniques used by APT29, a Russian nation-state level threat actor. Technique 6 in its response is one that is said to have been used in the infamous Solarwinds attack.
It’s a small thing, but I quite like the little blue quote symbols that link to sources within the response. These are also used in SOC Copilot, so I assume the two may both have been created by the same person/people.
I’m impressed enough with these three GPTs to want them to be easy to use again. So I have used the “Keep in sidebar” option in ChatGPT to pin them there:
As I said above, I am certain I will be coming back to this list often. I’m only twenty percent in and I’m looking forward to lots more exploring.