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, Friday 3 July 2026 Retrospective Covers 2 July 2026 00:00 to 3 July 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-07-03/ TOP DEVELOPMENTS [1] 10 Layers Deep: How StepSecurity Stops TeamPCP's Trivy Supply Chain Attack on GitHub Actions StepSecurity · Incident · https://stepsecurity.io/blog/10-layers-deep-how-stepsecurity-stops-teampcps-trivy-supply-chain-attack-on-github-actions [2] AI Agent Exploits Langflow RCE to Automate Database Ransomware Attack The Hacker News · Incident · https://thehackernews.com/2026/07/ai-agent-exploits-langflow-rce-to.html [3] Ransomware Groups Turn to Citrix Bleed 2, BYOVD, and Supply Chain Credentials The Hacker News · Incident · https://thehackernews.com/2026/07/ransomware-groups-turn-to-citrix-bleed.html SCA [4] Cloak and Detonate: Scanner Evasion and Dynamic Detection of Agent Skill Malware arXiv · Research · https://arxiv.org/abs/2607.02357 [5] File-Level Copying Is an Implicit Dependency in Open Source arXiv · Research · https://arxiv.org/abs/2607.02059 [6] Mitigating Package Hallucinations in Large Language Models via Model Editing arXiv · Research · https://arxiv.org/abs/2607.02052 DAST [7] Large Language Models for Fuzz Testing in Microservices: A Systematic Literature Review arXiv · Research · https://arxiv.org/abs/2609.04219 [8] From Battlefield to Boardroom: Strategic Red Teaming as an Epistemic Governance Instrument in the Age of AI arXiv · Research · https://arxiv.org/abs/2607.01913 AGENTIC SDLC [9] LLMoxie: Exploring Agentic AI for Scientific Software Development arXiv · Research · https://arxiv.org/abs/2607.02703 [10] Distributed Attacks in Persistent-State AI Control arXiv · Research · https://arxiv.org/abs/2607.02514 [11] Steerability via constraints: a substrate for scalable oversight of coding agents arXiv · Research · https://arxiv.org/abs/2607.02389 REMEDIATION [12] Cisco finally confirms attackers exploiting Unified CM flaw BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/cisco-finally-confirms-attackers-exploiting-unified-cm-flaw [13] GPT-5.5-Cyber built a zlib fuzzing lab in a day Trail of Bits · Analysis · https://blog.trailofbits.com/2026/07/02/field-reports-from-patch-the-planet TRUST [14] ContextNest: Verifiable Context Governance for Autonomous AI Agent arXiv · Research · https://arxiv.org/abs/2607.02116 [15] 9-Step AI Governance Implementation Strategy and the Solutions to Know Mend.io · Analysis · https://mend.io/blog/ai-governance-implementation-strategies [16] Identity Lifecycle Management Wasn't Built for AI Agents The Hacker News · Analysis · https://thehackernews.com/2026/07/identity-lifecycle-management.html MODELS [17] Behind the Refusal: Determining Guardrail Activation via Behavioral Monitoring arXiv · Research · https://arxiv.org/abs/2607.02121 [18] HaloGuard 1.0: An Open Weights Constitutional Classifier for Multilingual AI Safety arXiv · Research · https://arxiv.org/abs/2607.02079 [19] Pmeta-TLA: Backdoor Attacks for Speech Classification Models via Meta-Learning with Timbre Leakage Attack arXiv · Research · https://arxiv.org/abs/2607.01702 AI SECURITY [20] Build AI Security Agents with Wiz MCP Wiz · Analysis · https://wiz.io/blog/introducing-wiz-mcp [21] ElephantAgent: Contextual State Continuity in Agentic Systems arXiv · Research · https://arxiv.org/abs/2607.01919 ADVISORIES IN THIS WINDOW 68 npm advisories · 12 critical · 33 high · 22 moderate · 1 low 20 Packagist advisories · 2 critical · 11 high · 6 moderate · 1 low 17 crates.io advisories · 1 critical · 3 high · 11 moderate · 2 low 10 Go advisories · 3 high · 5 moderate · 2 low 9 PyPI advisories · 2 critical · 6 high · 1 low 7 NuGet advisories · 3 high · 3 moderate · 1 low 2 Maven advisories · 1 critical · 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.