Pierre Lalet

In our latest article, Baptiste MAYAUD describes the exploitation of CVE-2025-23016 affecting the FastCGI library, following an internal research project. The flaw is based on an integer overflow vulnerability leading to a buffer overflow on the heap, allowing arbitrary code execution under certain conditions. This bug has been present since the first versions of the library. The article details how the vulnerability works, the steps leading to exploitation and a full demonstration of the exploit. If you use FastCGI for your lightweight web server running in a 32-bit context, we recommend that you update the library to version 2.4.5. https://lnkd.in/eG2iT5Dz

39

3 commentaires

XposedOrNot += Association Nationale des Premiers Secours (ANPS) Data Breach The Association Nationale des Premiers Secours (ANPS) #databreach occurred in January 2026 when data was posted to a hacking forum, exposing 5.6K unique email addresses and associated personal information. Exposed data: Email addresses, Names, Dates of birth, Places of birth Potential risks: Identity theft, Phishing, Targeted scams, Privacy breaches

1

Today we are starting to roll out something new: Analyst Defense Labs (ADL) 👀 One thing has become very clear over the years — strong analysts don’t live in just one domain. Beyond deep Windows forensics and DFIR work, the day-to-day reality involves web compromises, detections, logs, cloud artifacts, tooling, and a lot of investigative tradecraft across different areas. Analyst Defense Labs is our answer to that. It’s a growing collection of focused micro-courses and guided investigation labs designed to help analysts continuously build and reinforce practical skills — one lab at a time. ADL complements our 101 Enterprise Security Fundamentals course and will become part of the Analyst Core track. Who do you get access? As a soft launch, ADL is available free for Hero users for the next 90 days, starting today. If you’re a Hero user, you’ll already see it in your dashboard. Access will open more broadly later on. This is just the beginning. 🚀

62

1 commentaire

🔥 New Security Alert: CVE-2025-67433 🔥 A critical heap overflow vulnerability has been discovered in Open TFTP Server, allowing unauthenticated attackers to crash the service — and in some cases potentially take control of systems running it. This isn’t theoretical — practical proof-of-concept exploitation exists, and misconfigured servers are at risk right now. 👉 Read the full vulnerability breakdown, impact analysis, and defensive guidance here: 🔗 https://lnkd.in/ggExrMKy #CyberSecurity #Infosec #VulnerabilityManagement #CVE202567433 #TFTP #ZeroDay #ThreatIntel #SecurityAwareness #NetworkSecurity #RiskManagement #SecurityOps #CyberThreats

2

🚨Cyber Attack Alert‼️ 🇫🇷France - CNFPT (Centre National de la Fonction Publique Territoriale) The CNFPT (Centre National de la Fonction Publique Territoriale) in France has suffered a cyberattack targeting its platform for external contributors (intervenant[.cnfpt[.fr). The attack potentially exposed personal data linked to user accounts, possibly including full datasets. While CNFPT confirmed the incident, it did not specify the exact data types affected. Reports indicate that hundreds of thousands of public agents were alerted. Discover more at hackrisk.io

2

The YesWeHack Report 2026 is here 🚀 Designed for security teams and our hacker community alike, this edition goes beyond a 2025 recap, it’s a blueprint for the future of offensive security. 🗺️ Discover our Map → Test → Fix → Comply vision, and see how this four-step cycle helps teams move from exposure to compliance with confidence. 🤖 Take a closer look at the role of AI in cybersecurity, including where automation truly adds value and where human expertise remains essential. 🛠️ Go behind the scenes with our customer success and triage teams, and learn how they support successful Bug Bounty Programs. 🏆 Meet this year’s Hall of Fame, explore our top-performing hunters, and revisit the milestones that marked 2025. Download the full report here: https://lnkd.in/eHdeEzyY

69

Ask ChatGPT: What is a pnhsm ? A pNHSM stands for personal (or portable) Network Hardware Security Module — sometimes also referred to as a private Network HSM, depending on usage. In simple terms: A pNHSM is a network-connected hardware security module that provides secure cryptographic services (key storage, encryption, signing, authentication) while ensuring that secret keys never leave the secure hardware. What does a pNHSM do? A pNHSM is used to: securely store cryptographic keys perform encryption, decryption, and digital signatures provide strong authentication, often for TLS connections replace weak mechanisms (passwords or PINs) with hardware-protected secrets (e.g. 256-bit TLS Pre-Shared Keys) How is it different from a traditional HSM? Traditional HSMpNHSMLarge, expensive, proprietaryCompact, affordableData-center-onlyDeployable anywhere (device or grid)Often closed interfacesUses standard TLSHigh power consumptionLow power consumption Key characteristics (Ethertrust context) In the Ethertrust approach: the pNHSM runs its own TLS server authentication is based on a 256-bit TLS Pre-Shared Key available as: secure-element grids for data centers or a small standalone device (“LeMonolith”) built on open hardware and software architectures (open-source, Arduino, Raspberry Pi, JavaCard) One-line definition 👉 A pNHSM is a lightweight, network-based HSM that delivers strong, hardware-rooted security using standard TLS, at lower cost and higher flexibility than traditional HSMs. https://lnkd.in/eDwyfu2q

11

3 commentaires