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, Monday 18 May 2026 Retrospective Covers 17 May 2026 00:00 to 18 May 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-05-18/ TOP DEVELOPMENTS [1] New Windows 'MiniPlasma' zero-day exploit gives SYSTEM access, PoC released BleepingComputer · Research · https://bleepingcomputer.com/news/microsoft/new-windows-miniplasma-zero-day-exploit-gives-system-access-poc-released [2] NGINX CVE-2026-42945 Exploited in the Wild, Causing Worker Crashes and Possible RCE The Hacker News · Incident · https://thehackernews.com/2026/05/nginx-cve-2026-42945-exploited-in-wild.html [3] Grafana GitHub Token Breach Led to Codebase Download and Extortion Attempt The Hacker News · Incident · https://thehackernews.com/2026/05/grafana-github-token-breach-led-to.html SCA [4] When Efficiency Backfires: Cascading LLMs Trigger Cascade Failure under Adversarial Attack arXiv · Research · https://arxiv.org/abs/2605.17288 AGENTIC SDLC [5] Rethinking Code Review in the Age of AI: A Vision for Agentic Code Review arXiv · Research · https://arxiv.org/abs/2605.17548 [6] MemRepair: Hierarchical Memory for Agentic Repository-Level Vulnerability Repair arXiv · Research · https://arxiv.org/abs/2605.17444 REMEDIATION [7] ContraFix: Skill-Enhanced Contrastive Runtime Analysis for Vulnerability Repair arXiv · Research · https://arxiv.org/abs/2605.17450 TRUST [8] Runtime Governance for Agentic AI: Action-Boundary Control with Trusted Provenance and Fail-Closed Execution arXiv · Research · https://arxiv.org/abs/2608.16891 MODELS [9] Fast and Lightweight Backdoor Detection via Head Random Probing arXiv · Research · https://arxiv.org/abs/2605.18908 AI SECURITY [10] AI Agents May Always Fall for Prompt Injections arXiv · Research · https://arxiv.org/abs/2605.17634 [11] ADR: An Agentic Detection System for Enterprise Agentic AI Security arXiv · Research · https://arxiv.org/abs/2605.17380 [12] ASPI: Seeking Ambiguity Clarification Amplifies Prompt Injection Vulnerability in LLM Agents arXiv · Research · https://arxiv.org/abs/2605.17324 ADVISORIES IN THIS WINDOW 1 Maven advisory · 1 moderate 1 PyPI advisory · 1 low 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.