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 4 July 2026 Retrospective Covers 3 July 2026 00:00 to 4 July 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-07-04/ TOP DEVELOPMENTS [1] Macro-Prudential AI Governance: A Two-Layer Early Warning and Response System for Frontier AI arXiv · Research · https://arxiv.org/abs/2607.03542 [2] AGL-1: The Enterprise AI Governance Layer as a Control Plane for Trusted Enterprise Intelligence arXiv · Research · https://arxiv.org/abs/2607.03516 [3] CAGE-1: Control, Assurance, and Governance Evaluation for Enterprise Agentic AI arXiv · Research · https://arxiv.org/abs/2607.03510 SCA [4] LLM-Enhanced Hierarchical Heterogeneous Graph Representation Learning for Malicious Python Package Detection arXiv · Research · https://arxiv.org/abs/2607.03350 [5] North Korea-Linked npm Packages Mimic Rollup Polyfills to Steal Developer Secrets The Hacker News · Analysis · https://thehackernews.com/2026/07/north-korea-linked-npm-packages-mimic.html DAST [6] Execution Divergence Graphs:Effective Discovery of Control-Flows from Execution Traces as Fuzzing Feedback arXiv · Research · https://arxiv.org/abs/2607.03396 AGENTIC SDLC [7] MOSAIC: Knowledge-Guided CLI Command Composition Attack in LLM Coding Agents arXiv · Research · https://arxiv.org/abs/2607.02857 TRUST [8] AI Systems as Digital Public Goods -- Evidence and Recommendations from a Multi-Stakeholder Assessment arXiv · Research · https://arxiv.org/abs/2607.03427 [9] The Foreign Policy AI Evaluation Gap arXiv · Research · https://arxiv.org/abs/2607.02955 MODELS [10] Overloading Large Vision-Language Models for Jailbreaking arXiv · Research · https://arxiv.org/abs/2607.02961 [11] TIER: Trajectory-Invariant Explanation Regularization for Membership Privacy arXiv · Research · https://arxiv.org/abs/2607.02903 AI SECURITY [12] Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies arXiv · Research · https://arxiv.org/abs/2607.03423 [13] CrackedPDFs: A Controlled Benchmark for Hidden Prompt Injection in PDFs arXiv · Research · https://arxiv.org/abs/2607.19396 ADVISORIES IN THIS WINDOW 15 Go advisories · 6 critical · 6 high · 3 moderate 14 npm advisories · 14 critical 1 PyPI 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.