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 15 October 2026 Retrospective Covers 14 October 2026 00:00 to 15 October 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-10-15/ TOP DEVELOPMENTS [1] Palo Alto Networks warns of PAN-OS RCE flaw exploited in attacks BleepingComputer · Incident · https://bleepingcomputer.com/news/security/palo-alto-networks-warns-of-pan-os-rce-flaw-exploited-in-attacks [2] Ledger asks sellers across Asia to halt sales after supply-chain attack BleepingComputer · Incident · https://bleepingcomputer.com/news/security/ledger-asks-sellers-across-asia-to-halt-sales-after-supply-chain-attack [3] curl Patches HTTP/2 Upload Flaw After PoC Demonstrates Client-Side Code Execution The Hacker News · Advisory · https://thehackernews.com/2026/10/curl-patches-http2-upload-flaw-after.html [4] Researchers Used a Public OpenAI Model to Build a $250,000 Chrome Full-Chain Exploit in Six Days The Hacker News · Research · https://thehackernews.com/2026/10/researchers-used-public-openai-model-to.html SAST [5] FreeSBee: A Constant-Time Compilation Pipeline arXiv · Research · https://arxiv.org/abs/2610.20171 [6] Mission (Im)Possible: Towards Autonomous Discovery of Autopilot Logic Flaws in Open-Source Drone Frameworks arXiv · Research · https://arxiv.org/abs/2610.19566 [7] Penetration testing every deployment with AWS Continuum AWS Security Blog · Analysis · https://aws.amazon.com/blogs/security/penetration-testing-every-deployment-with-aws-continuum SCA [8] CoRedSkill: Stealthy Skill Composition Attacks via AST-Guided Injection arXiv · Research · https://arxiv.org/abs/2610.20049 [9] Package managers have an integrity problem Trail of Bits · Analysis · https://blog.trailofbits.com/2026/10/14/package-managers-have-an-integrity-problem [10] Learn how Endor Labs detects malicious VS Code extensions Endor Labs · Analysis · https://endorlabs.com/learn/introducing-package-firewall-for-vs-code-extensions-detect-and-block-malicious-extensions-from-the-ide DAST [11] The Blind Spots of Agentic Pentesting and How to Close Them BleepingComputer · Guide · https://bleepingcomputer.com/news/security/the-blind-spots-of-agentic-pentesting-and-how-to-close-them [12] An AI Agent Found a Critical OTP Bypass in Production. The Attack Was the Easy Part. The Hacker News · Analysis · https://thehackernews.com/2026/10/an-ai-agent-found-critical-otp-bypass.html AGENTIC SDLC [13] Screwtape: Adversarial Transcript Editing for Monitor Red-Teaming arXiv · Research · https://arxiv.org/abs/2610.20941 [14] Thinking in Graphs: Neo4j-Backed Exploit-Chain Reasoning for LLM Penetration-Testing Agents arXiv · Research · https://arxiv.org/abs/2610.20902 [15] P2Patch: Exploit-Guided Vulnerability Repair arXiv · Research · https://arxiv.org/abs/2610.20245 REMEDIATION [16] VU#948176: Multiple TLS libraries can bypass delegation enforcement CERT/CC · Analysis · https://kb.cert.org/vuls/id/948176 [17] Microsoft fixes VPN issues caused by September Windows updates BleepingComputer · Analysis · https://bleepingcomputer.com/news/microsoft/microsoft-fixes-vpn-issues-caused-by-september-windows-updates MODELS [18] Counterfactual Neighborhood Risk Fields: Evidence-Grounded Selective Safety for Vision-Language Models arXiv · Research · https://arxiv.org/abs/2610.20679 [19] GeoCloak: An Agentic Framework for Semantics-Preserving Geolocation Privacy via Peripheral Clue Manipulation arXiv · Research · https://arxiv.org/abs/2610.20055 [20] Diagnose Before You Refine: Multi-Dimensional Judge Feedback for Query-Efficient Jailbreak Attacks arXiv · Research · https://arxiv.org/abs/2610.19971 AI SECURITY [21] From Detection to Decision: Calibration and Deferral for Prompt-Injection Detection arXiv · Research · https://arxiv.org/abs/2610.20907 [22] INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems arXiv · Research · https://arxiv.org/abs/2610.20776 [23] What Should a Software Team Spend on Agents? Simulation-Based Cost Optimization of Mixed AI-Human Teams arXiv · Research · https://arxiv.org/abs/2610.20394 ADVISORIES IN THIS WINDOW 16 npm advisories · 16 critical 4 PyPI advisories · 2 critical · 2 low 2 NuGet advisories · 1 high · 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.