I've decided to finally release my vulnerability hunting methodology that I've been using for several months now, both in testing as well as live engagements with great success.
I was inspired by (and huge congrats to) the @Blackfrost_AI team open-sourcing Cyber-Frost Harness, their reproducible security-agent runtime with procedural skills, an evidence ledger, and isolated Docker targets for red/blue/purple work. 🎉
If you like the Cyber Frost Harness, and want some out of the box vulnerabilty hunting and POC development kit to go along with it - check out my VulnHunter.
It's already loadable: their harness reads standard SKILL.md skills. Point skill_dir at VulnHunter's (or run its installer), and the hunting, falsification, and PoC disciplines load immediately.
Repo: https://t.co/tK2fhcxwrI
CREDITS: This is fork of @CapitalOne original repo. I simply took it a step further on POC development and model agnostic. I'm also actively developing in my forked repo as I make more changes to it and welcome the contributions from the community as well.
I think I'm doing the whole "respect who you give credit to" drama, and I asked my model to double and triple check it. So i can blame the model if its wrong. Otherwise, dont put me on blast on X if i suck at credits. /s
Looking forward to developing on this further with the community.
@DanielMiessler Yes it was a good read that took awhile too. Definitely things to think about when it comes to regulating AI. My take away was still glad I got the storage to grab local LLMs awhile I can. People are bad, but blanket banning access to knowledge is wrong!
@DanielMiessler Never left. Recognized the echo chamber for what it was. They can leave, they can come back. Doesn't matter. Here for the tech info and the memes.
@it_unprofession Yeah, fake it till you make! 💯 At 18 with 33 years experience life taught me this is how you get 💩 done in IT! I don't know everything and I am not afraid to look it up! Step by step if necessary!
@CristiVlad25 I've heard it and repeat it often that AI won't replace me, but people who know how to leverage AI will. So doing my best to upskill and stay knowledgeable and hands on with AI as much as I can.
@GoogleDeepMind For those interested, I asked the model itself and was informed I need download manually. There are no checks in the background to see if updates are available. It's on the user to go to huggingface.
@GoogleDeepMind For those of us running the Preview version, will the updates get pushed out via Google AI Edge Gallery or do we have to manually update? 🤔
🤓 OpenAI recently published an interesting threat report about how threat actors are using generative AI.
Here is a quick rundown: 👇
- Fake Jobs: IT workers using AI for deceptive hiring schemes.
- China IO: “Sneer Review” creates fake reviews and posts.
- Philippines IO: “High Five” floods political content.
- China IO: “VAGue Focus” targets US and EU via social engineering.
- Russia Cyber: “ScopeCreep” used GPT to write Go-based malware.
- China Cyber: “Vixen” & “Keyhole Panda” support full attack chains.
- China IO: “Uncle Spam” stirs division in US discourse.
- Iran IO: “STORM-2035” pushes geopolitical narratives in multiple languages.
- Cambodia Scam: “Wrong Number” runs AI-powered job scams for fake companies.
There is more TTPs in the report, and don't forget you can leverage NOVA to hunt for these patterns in your own system.
✨ https://t.co/g2LATRlord
@GoogleAI The models available in Google AI Edge Gallery appear to be a little enthusiastic in their responses. Were they made/trained at the time when all models experienced that euphoria?
@Havoc_XXI@Amer1can_Barbie Pretty much this. After qual, NBC didn't want to return with any tablets so would do one last run with any volunteers and they'd burn all remaining tablets. We'd go in and take our masks off to see who would last the longest. Good times.
“Go learn AI. Beat it up. Figure out how it works.”
AI is coming for dev jobs — cyber might be next.
Put yourself on the right side of this wave.
#NealBridges#CyberInsecurity#AIinCyber
@cyb3rops Well said. Over time I've learned how to prompt better. Still learning. And I do try different models to see how their responses compare. I wholeheartedly agree that we won't be replaced by AI, but rather by those that know how to use AI.
@GreenoneOld@TheCountofS They are not zombies. It's a rage virus! Go rewatch the first one. Not sure how they eat, but infected with "rage" does not equal "undead".
🚨 Reports suggest US authorities may ban TP-Link Wi-Fi routers in 2025. Regulated industries, ensure your end users aren't connected to TP-Link routers. Use MDE discovery and DefenderXDR's SeenBy() to detect connections. 🛡️📡