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 Daily, Wednesday 9 September 2026 Retrospective Covers 8 September 2026 00:00 to 9 September 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-09-09/ TOP DEVELOPMENTS [1] Microsoft September 2026 Patch Tuesday fixes 966 flaws, 2 zero-days BleepingComputer · Advisory · https://bleepingcomputer.com/news/microsoft/microsoft-september-2026-patch-tuesday-fixes-966-flaws-2-zero-days [2] Microsoft releases Windows 10 KB5122878 extended security update BleepingComputer · Research · https://bleepingcomputer.com/news/microsoft/microsoft-releases-windows-10-kb5122878-extended-security-update [3] CVE-2026-75650: Adobe Commerce and Magento Improper Neutralization of Special Elements Used in a Template Engine Vulnerability added to CISA's Known Exploited Vulnerabilities catalog CISA · Advisory · https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2026-75650 SAST [4] LLMSec-AV: A Vulnerability Taxonomy and LLM-Driven Software Weakness Discovery Framework for Autonomous Vehicles arXiv · Research · https://arxiv.org/abs/2609.09386 [5] Testing race conditions with memory access tracing and stack-based delay injection Google · Analysis · https://projectzero.google/2026/09/maccconc-race-condition.html SCA [6] Measuring the Security of the Evolving Software Supply Chain: a Research Agenda arXiv · Research · https://arxiv.org/abs/2609.08810 DAST [7] Why federal cyber defense demands an offense-driven mindset CyberScoop · Analysis · https://cyberscoop.com/offense-driven-federal-cyber-defense [8] LLM-Based Penetration Testing in the Presence of Honeypots arXiv · Research · https://arxiv.org/abs/2609.08093 [9] Jason Haddix: Stop fearing AI pentesting Aikido Security · Analysis · https://aikido.dev/blog/stop-fearing-ai-pentesting AGENTIC SDLC [10] Authority Is Not a String: A Capability-Scoped Harness for Prompt-Injection-Resistant Coding Agents arXiv · Research · https://arxiv.org/abs/2609.08371 [11] SWE-Bench Pro Verified: A Reliable Benchmark for Software Engineering Agents arXiv · Research · https://arxiv.org/abs/2609.08149 [12] 1Password increases engineering productivity 21% with Codex OpenAI News · Analysis · https://openai.com/index/1password REMEDIATION [13] Microsoft Plugs Nearly 1,000 Security Holes Krebs on Security · Analysis · https://krebsonsecurity.com/2026/09/microsoft-plugs-nearly-1000-security-holes [14] Why this month's Microsoft patch release is a doozy Ars Technica: Security · Release · https://arstechnica.com/security/2026/09/microsoft-patches-a-record-972-vulnerabilities-112-of-them-critical [15] Windows 11 cumulative updates KB5124008 & KB5122880 released BleepingComputer · Analysis · https://bleepingcomputer.com/news/microsoft/windows-11-cumulative-updates-kb5124008-and-kb5122880-released TRUST [16] The EU CRA's Real Question: What Shipped, and When Did You Know? BleepingComputer · Standard / policy · https://bleepingcomputer.com/news/security/the-eu-cras-real-question-what-shipped-and-when-did-you-know [17] Why 2026 is the Year to Upgrade to an Agentic AI SOC Elastic · Analysis · https://elastic.co/security-labs/blog/why-2026-is-the-year-to-upgrade-to-an-agentic-ai-soc [18] Seeing is Not Believing: Breaking the Physical-to-Digital Trust Boundary in Robotics arXiv · Research · https://arxiv.org/abs/2609.08280 MODELS [19] DuplexJail: Spoken Interruption Attacks on Full-Duplex Speech Models arXiv · Research · https://arxiv.org/abs/2609.09420 [20] Understanding In-Context Multimodal Jailbreaks via Posterior Reweighting arXiv · Research · https://arxiv.org/abs/2609.10613 AI SECURITY [21] An Experimental Evaluation of Multimodal Prompt Injection Attacks on Agentic AI Frameworks arXiv · Research · https://arxiv.org/abs/2609.09404 [22] MemSentry: A Framework for Detecting Persistent Memory Poisoning in Agentic AI arXiv · Research · https://arxiv.org/abs/2609.08747 [23] Hackers build AI frameworks for widescale credential theft BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/hackers-build-ai-frameworks-for-widescale-credential-theft ADVISORIES IN THIS WINDOW 52 npm advisories · 13 critical · 23 high · 13 moderate · 3 low 30 PyPI advisories · 5 critical · 15 high · 10 moderate 9 Go advisories · 1 critical · 2 high · 5 moderate · 1 low 7 Packagist advisories · 2 critical · 4 high · 1 moderate 3 NuGet advisories · 1 critical · 1 high · 1 moderate 2 Maven advisories · 1 critical · 1 moderate 1 Multi-ecosystem 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.