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 Weekly, week 31 of 2026 Retrospective Covers 21 July 2026 11:00 to 28 July 2026 11:00 UTC Web: https://keryn.dev/headlines/weekly/2026-W31/ TOP DEVELOPMENTS [1] Check Point warns of SmartConsole zero-day exploited in attacks BleepingComputer · Incident · https://bleepingcomputer.com/news/security/check-point-patches-smartconsole-zero-day-exploited-in-attacks [2] Arista patches VeloCloud Orchestrator zero-day exploited in attacks BleepingComputer · Advisory · https://bleepingcomputer.com/news/security/arista-patches-velocloud-orchestrator-zero-day-exploited-in-attacks [3] CISA orders urgent action on actively exploited Langflow RCE flaw BleepingComputer · Incident · https://bleepingcomputer.com/news/security/cisa-orders-feds-to-patch-actively-exploited-langflow-rce-flaw [4] Critical SharePoint RCE flaw exploited to steal machine keys BleepingComputer · Incident · https://bleepingcomputer.com/news/security/critical-sharepoint-rce-flaw-exploited-to-steal-machine-keys [5] CVE-2025-68686: Fortinet FortiOS Exposure of Sensitive Information to an Unauthorized Actor Vulnerability added to CISA's Known Exploited Vulnerabilities catalog CISA · Advisory · https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2025-68686 SAST [6] Hybrid Analysis for Secure MCP Tool Use in LLM Agents arXiv · Research · https://arxiv.org/abs/2607.25297 [7] SecDrift: Measuring Sector-Conditioned Security Drift in AI-Generated Code arXiv · Research · https://arxiv.org/abs/2607.25225 [8] ZIMPAF & RedPhuzz: High-fidelity Web Application Fuzzing via Branch, Language Construct, and Function Call Monitoring arXiv · Research · https://arxiv.org/abs/2607.25012 [9] ALIBI: Adaptive Agentic Attacks on LLM-Based Vulnerability Detectors via Adversarial Code Comments arXiv · Research · https://arxiv.org/abs/2607.24964 SCA [10] Compromised PyPI Package: mrmustard 0.7.4 Steals SSH, Cloud, and Kubernetes Credentials StepSecurity · Incident · https://stepsecurity.io/blog/compromised-pypi-mrmustard-0-7-4-credential-stealer [11] Trusting-Trust Attack against an Entire Linux Distribution through Binary Manipulation arXiv · Research · https://arxiv.org/abs/2607.24888 [12] GitHub, PyPI add time-based defenses against supply chain attacks BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/github-pypi-add-time-absed-defenses-against-supply-chain-attacks [13] No Edges, No Verdict: A Large-Scale Empirical Study of Declared Dependency Graphs in 78K SBOMs in the Wild arXiv · Research · https://arxiv.org/abs/2607.22140 DAST [14] Modeling Local Exploit Hazard - A Bayesian Framework for Quantifying Exploit Risk and Operational Efficiency arXiv · Research · https://arxiv.org/abs/2607.24618 [15] The Illusion of Secure LLM Code: Closing the Security Gap via Iterative Reprompting arXiv · Research · https://arxiv.org/abs/2607.23710 [16] NodeBB Patches Eight AI-Found Flaws Exposing Admin Access and Private Chats The Hacker News · Advisory · https://thehackernews.com/2026/07/nodebb-patches-eight-ai-found-flaws.html [17] Quoting Thomas Ptacek Simon Willison · Analysis · https://simonwillison.net/2026/Jul/22/thomas-ptacek AGENTIC SDLC [18] Microsoft Azure DevOps MCP Flaw Lets Hidden PR Comments Hijack AI Review Agents The Hacker News · Incident · https://thehackernews.com/2026/07/microsoft-azure-devops-mcp-flaw-lets.html [19] NTT DATA Group cuts incident analysis to 30 minutes with Codex OpenAI News · Incident · https://openai.com/index/ntt-data [20] Where Is the Cost of Third-Party API Routers in Agentic Software Development? arXiv · Research · https://arxiv.org/abs/2607.23624 [21] Slopsquatting, Phantom Domains, and HalluSquatting Are the Same AI Attack BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/slopsquatting-phantom-domains-and-hallusquatting-are-the-same-ai-attack REMEDIATION [22] CVE-2026-50522: Microsoft SharePoint Deserialization of Untrusted Data Vulnerability added to CISA's Known Exploited Vulnerabilities catalog CISA · Advisory · https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2026-50522 [23] Hackers target US firms in FastJson RCE zero-day attacks BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/hackers-target-us-firms-in-fastjson-rce-zero-day-attacks [24] Russian hackers exploit Zimbra zero-click flaw for email theft BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/russian-hackers-exploit-zimbra-zero-click-flaw-for-email-theft [25] Adobe Chrome extension flaw let sites access private WhatsApp chats BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/adobe-chrome-extension-flaw-let-sites-access-private-whatsapp-chats TRUST [26] ARCHER: Agentic Rule and Compliance Harness for Executable Regulations arXiv · Research · https://arxiv.org/abs/2607.25566 [27] Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering arXiv · Research · https://arxiv.org/abs/2607.25479 [28] TRACE-CTI: Auditable Post-Extraction Governance of TTP Claims with Knowledge Graphs arXiv · Research · https://arxiv.org/abs/2607.24563 [29] Regulating for AI Legitimacy arXiv · Research · https://arxiv.org/abs/2607.24391 MODELS [30] Anti-Backdoor Coreset Selection via Cumulative Entropy arXiv · Research · https://arxiv.org/abs/2607.25502 [31] Lilith: Backdoor Generalization under Training-Inference Trigger Shift arXiv · Research · https://arxiv.org/abs/2607.26099 [32] Evolving Safety Landscape of Multi-modal Large Language Models: A Survey of Emerging Threats and Safeguards arXiv · Research · https://arxiv.org/abs/2608.07535 [33] ScoreShield: Differentially Private Release of Similarity Scores arXiv · Research · https://arxiv.org/abs/2607.25041 AI SECURITY [34] Hermes AI agent used to automate attack on Thai Finance Ministry BleepingComputer · Incident · https://bleepingcomputer.com/news/security/hermes-ai-agent-used-to-automate-attack-on-thai-finance-ministry [35] APPA: Recoverable Information-Flow Control for Real-World LLM Agents arXiv · Research · https://arxiv.org/abs/2607.24625 [36] Atlas: Wiz's autonomous AI Agent for vulnerability research, ranked #1 on CyberGym Wiz · Analysis · https://wiz.io/blog/atlas-ai-vulnerability-researcher [37] Just Testing, Move Along: Evasion of LLM-based System Log Interpretation by Prompt Injection arXiv · Research · https://arxiv.org/abs/2607.24174 ADVISORIES IN THIS WINDOW 11571 PyPI advisories · 11514 critical · 30 high · 19 moderate · 8 low 2937 npm advisories · 2830 critical · 61 high · 45 moderate · 1 low 75 Go advisories · 5 critical · 26 high · 38 moderate · 6 low 43 NuGet advisories · 2 critical · 13 high · 10 moderate · 18 low 41 Maven advisories · 5 critical · 18 high · 18 moderate 16 Packagist advisories · 2 critical · 7 high · 5 moderate · 2 low 8 crates.io advisories · 2 high · 5 moderate · 1 low 6 RubyGems advisories · 1 high · 2 moderate · 3 low 3 Multi-ecosystem advisories · 1 high · 1 moderate · 1 low 1 Swift advisory · 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.