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, Saturday 20 June 2026 Retrospective Covers 19 June 2026 00:00 to 20 June 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-06-20/ TOP DEVELOPMENTS [1] CISA: Splunk Enterprise flaw actively exploited, patch by Sunday BleepingComputer · Incident · https://bleepingcomputer.com/news/security/cisa-splunk-enterprise-flaw-actively-exploited-patch-by-sunday [2] AutoJack: How a single page can RCE the host running your AI agent Microsoft · Analysis · https://microsoft.com/en-us/security/blog/2026/06/18/autojack-single-page-rce-host-running-ai-agent [3] CISA warns Fortinet users to secure devices after FortiBleed leak BleepingComputer · Incident · https://bleepingcomputer.com/news/security/cisa-warns-fortinet-users-to-secure-devices-after-fortibleed-leak SAST [4] Evaluating LLMs for Real-World Web Vulnerability Detection arXiv · Research · https://arxiv.org/abs/2606.21397 [5] "What Happens Locally, Leaks Globally": Detecting Privacy Leakage Risks in MCP Servers arXiv · Research · https://arxiv.org/abs/2606.21338 SCA [6] A Longitudinal Study of Android Apps Signing Key Protection arXiv · Research · https://arxiv.org/abs/2606.21487 MODELS [7] Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs arXiv · Research · https://arxiv.org/abs/2606.21638 [8] Scalable Hierarchical Attention Transformers for Multi-Turn Jailbreak Detection in Long Conversations arXiv · Research · https://arxiv.org/abs/2606.21082 [9] OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization arXiv · Research · https://arxiv.org/abs/2606.21077 AI SECURITY [10] Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents arXiv · Research · https://arxiv.org/abs/2606.21732 [11] Every AI Agent Is an Identity. Most Organizations Don't Treat Them That Way BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/every-ai-agent-is-an-identity-most-organizations-dont-treat-them-that-way [12] AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI Agents arXiv · Research · https://arxiv.org/abs/2606.21129 ADVISORIES IN THIS WINDOW 46 npm advisories · 11 critical · 16 high · 12 moderate · 7 low 30 PyPI advisories · 1 critical · 18 high · 11 moderate 28 Go advisories · 2 critical · 9 high · 15 moderate · 2 low 20 Packagist advisories · 1 critical · 4 high · 12 moderate · 3 low 16 RubyGems advisories · 13 high · 1 moderate · 2 low 16 NuGet advisories · 2 critical · 6 high · 7 moderate · 1 low 10 Maven advisories · 1 critical · 3 high · 6 moderate 7 crates.io advisories · 1 high · 6 moderate 1 GitHub Actions advisory · 1 high 1 Multi-ecosystem advisory · 1 high 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.