HAIJA INTEL REPORT

New DFIR Labs case: A Contagious Interview, Private Case #38413 A routine 'hiring assessment' turns into an endpoint compromise when a developer is social-engineered into cloning a malicious GitHub repo. Trace the campaign through obfuscated Node.js, Python payloads, C2, host

Also, it can be a 1,000,000,000x ROI if you exploit a drain bug in a DeFi protocol only few people know this, the ROI for $200 gpt/claude subscription is 1000x in cyber security.

Rapid7 Labs recently uncovered a crypto scammer's working environment via a misconfigured web directory, exposing raw phone-number datasets, phishing panels, voice-dialing scripts, counterfeit wallet builds, and more. Full technical analysis here: https:// r-7.co/4xban7K

One compromise. A much bigger blast radius. Software Supply Chain Security for Red Teamers explores how attackers abuse trust across the software supply chain & the defenses designed to stop them. Start learning: https:// academy.specterops.io/software-suppl y-chain-security-for-

We just launched @HackingHub’s latest course: “ @CaidoIO for Hackers” Learn to maximize Caido: from intercepting and replaying traffic to fuzzing, workflows, plugins, and chaining real bugs. 18 modules. 52 lessons. Hands-on labs. Lifetime access. Enroll

Karanbir Singh, CrowdStrike VP of Product, Cloud Security, explores what’s new and what’s next for Falcon Cloud Security. As cloud-conscious adversaries evolve their tradecraft and AI reshapes the security landscape, he’ll share where the market is headed, how Falcon Cloud

Blueprint to Making $100K with @CaidoIO (Free Bug Bounty Course) with @amrelsagaei https:// youtu.be/DXIOQZpS25U

Business logic flaws are where AI security testing gets humbled. No CVE to match, no pattern to grep for, just workflows that behave badly in ways only context reveals. @ehrishiraj talks through what LLMs can and can't do here on the latest Application Security Weekly. Listen

Can we teach AI to reason through detection evidence before making a triage verdict? Our latest research says yes. We’ve developed a reasoning-enabled triage model that significantly outperforms general-purpose models when evaluating real-world Windows endpoint detections. By

Are your security tools missing threats hiding under the radar? Watch our latest video to learn how Cisco Talos Threat Hunting combines AI and expert human analysis to uncover sophisticated adversary activity that traditional detection methods often miss: https:// youtu.be/8yUXRI

Or even the AI companies doing cool Offsec work? Who are the team's you'd say are doing the coolest stuff in the AI RedTeam space at the minute? I'm struggling to keep track of where all the cool releases are coming from

Moved over to LangFuse for tracking my LLM traces, pretty nice!