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, Thursday 29 October 2026 Retrospective Covers 28 October 2026 00:00 to 29 October 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-10-29/ TOP DEVELOPMENTS [1] Hackers now scanning systems for recently-patched Oracle flaw BleepingComputer · Advisory · https://bleepingcomputer.com/news/security/hackers-now-scanning-systems-for-recently-patched-oracle-flaw [2] Adobe Fixes 8 Critical Campaign Classic Flaws Allowing Code Execution and Security Bypass The Hacker News · Incident · https://thehackernews.com/2026/10/adobe-fixes-8-critical-campaign-classic.html [3] ShinyHunters-Exploited Oracle PeopleSoft Zero-Day Turns Out to Be Three Chained Flaws The Hacker News · Incident · https://thehackernews.com/2026/10/shinyhunters-exploited-oracle.html SAST [4] How Deterministic SAST and AI Security Agents Divide the Work: Coverage, Harness Policy and Cost in Snyk VulnBench 2.0 arXiv · Research · https://arxiv.org/abs/2610.38617 [5] Endor Labs AI SAST Finds Two Apache Kafka Zero-Days: Unbounded Memory Allocation from Brokers to Kafka Streams Endor Labs · Analysis · https://endorlabs.com/learn/endor-labs-ai-sast-finds-two-apache-kafka-zero-days-unbounded-memory-allocation-from-brokers-to-kafka-streams SCA [6] Prioritize Dependabot alerts with reachability analysis GitHub · Analysis · https://github.blog/changelog/2026-10-28-prioritize-dependabot-alerts-with-reachability-analysis [7] This Shit is Hard: Nobody has ever been pwned by Helvetica…or have they? Chainguard · Analysis · https://chainguard.dev/unchained/this-shit-is-hard-nobody-has-ever-been-pwned-by-helvetica-or-have-they AGENTIC SDLC [8] CARI: Cross-Boundary Adversarial Runtime Injection in Coding-Agent Workflows arXiv · Research · https://arxiv.org/abs/2610.38799 [9] Security Impact of Coding Agent's Memory Files arXiv · Research · https://arxiv.org/abs/2610.38521 [10] The ADLC maturity model: 5 stages of agentic development security Endor Labs · Analysis · https://endorlabs.com/learn/adlc-maturity-model REMEDIATION [11] DeepDiff: Neuro-Dynamic Regression-Freedom Checking for Security Patch Validation arXiv · Research · https://arxiv.org/abs/2610.39066 [12] Tracekit Box: Running AI Agents in Micro-VMs Where Every Way Out Is Decided and Signed arXiv · Research · https://arxiv.org/abs/2610.38458 [13] The ‘Vulnpocalypse’ is coming for the internet’s aging code, says Wiz’s CTO TechCrunch: Security · Analysis · https://techcrunch.com/podcast/the-vulnpocalypse-is-coming-for-the-internets-aging-code-says-wizs-cto TRUST [14] ADLC compliance: Mapping agentic development to NIST SSDF, OWASP, and the EU AI Act Endor Labs · Standard / policy · https://endorlabs.com/learn/adlc-compliance [15] Snyk UNFILTERED: Inside Our First-Ever Customer Summit Snyk · Analysis · https://snyk.io/blog/snyk-unfiltered-customer-summit MODELS [16] FlipJev: Fast Black-Box Red Teaming Finds Universal Adversarial Suffixes for Typed Decision Models arXiv · Research · https://arxiv.org/abs/2610.38895 [17] Discovering Jailbreaks via Curriculum of Models arXiv · Research · https://arxiv.org/abs/2610.38833 [18] Rethinking LLM Jailbreak Evaluation: Continuous Risk Landscapes Beyond Binary Outcomes arXiv · Research · https://arxiv.org/abs/2610.38595 AI SECURITY [19] Where You Inject Matters: The Trust Region Effect in LLM Agents arXiv · Research · https://arxiv.org/abs/2610.38867 [20] Post-Training Large Language Models with LLMON arXiv · Research · https://arxiv.org/abs/2610.38120 ADVISORIES IN THIS WINDOW 57 npm advisories · 20 critical · 24 high · 10 moderate · 3 low 26 PyPI advisories · 2 critical · 2 high · 19 moderate · 3 low 14 Go advisories · 1 high · 12 moderate · 1 low 2 crates.io advisories · 2 critical 1 RubyGems advisory · 1 moderate 1 Multi-ecosystem advisory · 1 moderate 1 Maven advisory · 1 moderate 1 NuGet 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.