PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address. KERYN HEADLINES Weekly, week 18 of 2025 Retrospective Covers 22 April 2025 11:00 to 29 April 2025 11:00 UTC Web: https://keryn.dev/headlines/weekly/2025-W18/ TOP DEVELOPMENTS [1] CVE-2025-31324: SAP NetWeaver Unrestricted File Upload Vulnerability added to CISA's Known Exploited Vulnerabilities catalog CISA · Advisory · https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-31324 [2] Ripple's recommended XRP library xrpl.js hacked to steal wallets BleepingComputer · Incident · https://bleepingcomputer.com/news/security/ripples-recommended-xrp-library-xrpljs-hacked-to-steal-wallets [3] HiddenLayer Researchers Identify Novel Universal Bypass for All Major LLMs CoSAI · Research · https://coalitionforsecureai.org/hiddenlayer-researchers-identify-novel-universal-bypass-for-all-major-llms [4] Cookie-Bite attack PoC uses Chrome extension to steal session tokens BleepingComputer · Research · https://bleepingcomputer.com/news/security/cookie-bite-attack-poc-uses-chrome-extension-to-steal-session-tokens [5] GCP Cloud Composer Bug Let Attackers Elevate Access via Malicious PyPI Packages The Hacker News · Research · https://thehackernews.com/2025/04/gcp-cloud-composer-bug-let-attackers.html SAST [6] SoK: Cryptocurrency Mixing and Anonymity - Architectures, Threat Models, Operational Aspects and Security arXiv · Research · https://arxiv.org/abs/2504.20296 [7] LLMpatronous: Harnessing the Power of LLMs For Vulnerability Detection arXiv · Research · https://arxiv.org/abs/2504.18423 [8] Context-Enhanced Vulnerability Detection Based on Large Language Model arXiv · Research · https://arxiv.org/abs/2504.16877 [9] Neuro-symbolic Static Analysis with LLM-generated Vulnerability Patterns arXiv · Research · https://arxiv.org/abs/2504.16057 SCA [10] Sleeping Giants -- Activating Dormant Java Deserialization Gadget Chains through Stealthy Code Changes arXiv · Research · https://arxiv.org/abs/2504.20485 [11] Automatically Generating Rules of Malicious Software Packages via Large Language Model arXiv · Research · https://arxiv.org/abs/2504.17198 [12] Vulnerability Enumeration Conundrum – an Open Source Perspective on CVE and CWE OpenSSF · Analysis · https://openssf.org/blog/2025/04/23/vulnerability-enumeration-conundrum-an-open-source-perspective-on-cve-and-cwe [13] Implementing AI Bill of Materials (AI BOM) with SPDX 3.0: A Comprehensive Guide to Creating AI and Dataset Bill of Materials arXiv · Research · https://arxiv.org/abs/2504.16743 DAST [14] kAgent: An execution-guided crash resolution agent for the Linux kernel arXiv · Research · https://arxiv.org/abs/2504.20412 [15] Kali Linux warns of update failures after losing repo signing key BleepingComputer · Analysis · https://bleepingcomputer.com/news/linux/kali-linux-warns-of-update-failures-after-losing-repo-signing-key [16] Document My Pentest: you hack, the AI writes it up! PortSwigger Research · Analysis · https://portswigger.net/research/document-my-pentest AGENTIC SDLC [17] On Developers' Self-Declaration of AI-Generated Code: An Analysis of Practices arXiv · Research · https://arxiv.org/abs/2504.16485 [18] GitHub for Beginners: Building a REST API with Copilot GitHub · Analysis · https://github.blog/ai-and-ml/github-copilot/github-for-beginners-building-a-rest-api-with-copilot [19] Endor Labs, which builds tools to scan AI-generated code for vulnerabilities, lands $93M TechCrunch: Security · Analysis · https://techcrunch.com/2025/04/23/endor-labs-which-builds-tools-to-scan-ai-generated-code-for-vulnerabilities-lands-93m REMEDIATION [20] Windows "inetpub" security fix can be abused to block future updates BleepingComputer · Analysis · https://bleepingcomputer.com/news/microsoft/windows-inetpub-security-fix-can-be-abused-to-block-future-updates [21] ASUS releases fix for AMI bug that lets hackers brick servers BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/asus-releases-fix-for-ami-bug-that-lets-hackers-brick-servers TRUST [22] SAGA: A Security Architecture for Governing AI Agentic Systems arXiv · Research · https://arxiv.org/abs/2504.21034 [23] NRevisit: A Cognitive Behavioral Metric for Code Understandability Assessment arXiv · Research · https://arxiv.org/abs/2504.18345 [24] The Malicious Technical Ecosystem: Exposing Limitations in Technical Governance of AI-Generated Non-Consensual Intimate Images of Adults arXiv · Research · https://arxiv.org/abs/2504.17663 [25] How Effective are Generative Large Language Models in Performing Requirements Classification? arXiv · Research · https://arxiv.org/abs/2504.16768 MODELS [26] FFCBA: Feature-based Full-target Clean-label Backdoor Attacks arXiv · Research · https://arxiv.org/abs/2504.21054 [27] SFIBA: Spatial-based Full-target Invisible Backdoor Attacks arXiv · Research · https://arxiv.org/abs/2504.21052 [28] When Memory Becomes a Vulnerability: Towards Multi-turn Jailbreak Attacks against Text-to-Image Generation Systems arXiv · Research · https://arxiv.org/abs/2504.20376 [29] Does More Compute Make Every Adversarial Attack Transferable? arXiv · Research · https://arxiv.org/abs/2504.20310 AI SECURITY [30] Token-Efficient Prompt Injection Attack: Provoking Cessation in LLM Reasoning via Adaptive Token Compression arXiv · Research · https://arxiv.org/abs/2504.20493 [31] Robustness via Referencing: Defending against Prompt Injection Attacks by Referencing the Executed Instruction arXiv · Research · https://arxiv.org/abs/2504.20472 [32] Simplified and Secure MCP Gateways for Enterprise AI Integration arXiv · Research · https://arxiv.org/abs/2504.19997 [33] Securing Agentic AI: A Comprehensive Threat Model and Mitigation Framework for Generative AI Agents arXiv · Research · https://arxiv.org/abs/2504.19956 ADVISORIES IN THIS WINDOW 89 npm advisories · 80 critical · 4 high · 3 moderate · 2 low 18 Packagist advisories · 1 critical · 3 high · 11 moderate · 3 low 15 Maven advisories · 1 critical · 3 high · 9 moderate · 2 low 10 Go advisories · 1 critical · 2 high · 5 moderate · 2 low 5 PyPI advisories · 1 critical · 2 moderate · 2 low 2 NuGet advisories · 1 critical · 1 low 1 RubyGems advisory · 1 moderate All advisories: https://keryn.dev/headlines/advisories/ Titles, links and short excerpts only; every item links to its source. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.