Joaquín Molina

We have started releasing modules for the Agentic Cybersecurity AI and Python Development course in the Cyber Range, with new videos being added every day. This week we added about five more videos to the AI and Python section. We also have about eight more videos in editing, including three that introduce the next section, Python and AI Building Blocks: • Introduction to Advanced AI with Python • Installing the OpenAI Library with pip • Generating our OpenAI API Key A new category has also been added in both the Cyber Range and the Community. My goal is to help everyone stay ahead of the AI curve as much as possible, so I’ve been doing a lot of research and putting it into practice in this course. There is quite a bit of python, but don’t worry since you won’t need to build big stuff from scratch. I’ll walk you through everything, and you’ll just need to understand and interpret code. Will do our best to get these out as soon as possible! If you want to check it out, you can find it in the Cyber Range: https://lnkd.in/gB3RuSGA 

139

21 comentarios

‼️🚨Now Available: NIST SP 800-53 Release 5.2.0🚨‼️ Updates to the security and privacy control catalog and assessment procedures are out! 🆕 What’s New in Release 5.2.0? As directed by Executive Order 14306, this release focuses on improving the security and reliability of software updates and patches. Access the latest and greatest through the NIST Cybersecurity and Privacy Reference Tool, including in OSCAL! While SP 800-53 already covered many foundational practices, Release 5.2.0 provides additional supplemental guidance and new controls addressing: ✅ Resiliency by design ✅ Software development and testing best practices ✅ Software deployment and update management approaches ✅ Software integrity and validation ✅ Root cause analysis 🤓 Learn more and get Release 5.2.0: https://lnkd.in/et57Y6qd #nist #sp80053 #cybersecurity #privacy #cscrm #controls #assessment #patchmanagement #secureupdates #riskmanagement

361

8 comentarios

⚡ Two threat actors, similar strategy: large scale compromises are largely invisible to detection and response teams. 200+ U.S. organizations and 80 countries and counting... 📰 New reporting on Salt Typhoon's systematic targeting of telecom backbone infrastructure coincides with Silk Typhoon's escalating attacks on cloud service providers. Different attacks, same playbook: compromise the solutions provider once, access all their customers. Whether it's compromised backbone routers or trusted cloud providers, these attacks are largely invisible to corporate security teams. Time to revisit our detection and response strategies before it's too late. ➡️ The latest reporting on Salt Typhoon from the The Washington Post: https://lnkd.in/gCMr4GdC ➡️ Coverage of Silk Typhoon activity from CyberScoop: https://lnkd.in/gFbWg-_U 🔵 Cybersecurity leadership news and insights via Substack: https://lnkd.in/gtfQvqsn Push Security #cyberthreats #cybersecurity #cloudsecurity #securityoperations #incidentresponse #securityleadership

82

🔍 Public Directory Listing Exposed on target.com During routine reconnaissance, I discovered that the subdomain json-schema.target.com had an open directory listing enabled at the root path /. This exposed several internal schema folders such as: /app/ /data-class/ /form/ /ivy/ /market/ /openapi/ /openrpc/ /process/ /variables/ 🛡️ Security Risk: Public directory listing can unintentionally expose: Internal project structures Sensitive files or APIs Development or staging assets Unauthenticated access to schema blueprints This misconfiguration could aid attackers in identifying further attack surfaces or performing targeted exploitation. ✅ I’ve responsibly disclosed this to the target.com security team to help them remediate the issue. 🔐 #CyberSecurity #BugBounty #EthicalHacking #DirectoryListing #InfoSec #Reconnaissance #WebSecurity #ResponsibleDisclosure #OSINT #CyberAwareness

61

2 comentarios

Beelzebub: Open-source Honeypot! AI-powered honeypot framework creates a secure, high-interaction environment to detect and analyze cyber threats. It’s designed for low-code deployment and supports multi-protocol decoys like SSH, HTTP, TCP, and MCP to detect prompt injection attacks against LLM agents. • Low-code YAML configuration for rapid setup and scaling. • AI/LLM integration to simulate system behavior without actual risk. • Multi-protocol support: SSH, HTTP, TCP, MCP. • Observability: Prometheus metrics, ELK stack ready. • Deployment flexibility: Docker, Kubernetes (Helm), Go compiler. • Integrations: RabbitMQ, OpenMetrics. • High-interaction behavior with a safe low-interaction backend. Unlike traditional honeypots, Beelzebub uses an LLM sandbox to interact with attackers safely. Source: https://lnkd.in/gm47Wv6X Without data analysis, a honeypot is just a log generator! Discover over 10+ essential data analysis techniques for effective threat hunting in my "Cyber Threat Hunt 101" YouTube series, explained simply: https://lnkd.in/gkVB6B2j Please share and subscribe if you enjoy the content! #cybersecurity #threathunting #threatdetection #blueteam #soc #socanalyst #skillsdevelopment #careergrowth #IR #DataAnalysis #IncidentResponse

304

8 comentarios

Building trust in AI pentesting is not easy. Hackbots face ongoing challenges with false positives or missed vulnerabilities. In this blog, Pedro Conde explains how Ethiack research tackles these issues through rigorous testing, post-hoc reasoning and a little help from a specialized sub-agent called "Verifier". What's covered? ✅ How AI hackbots scale vulnerability discovery with human-like reasoning ✅ Why unbiased, benchmark-based evaluation is essential for reliability ✅ How the Verifier filters hallucinated vulnerabilities ✅ Examples where validation prevented false reports ✅ Comparison of Hackbot results vs. conventional scanners like ZAP Read the full post here: https://lnkd.in/dDXRCkE7 #AI #Pentesting #CyberSecurity #VulnerabilityTesting #OWASP

47