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 25 July 2026 Retrospective Covers 24 July 2026 00:00 to 25 July 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-07-25/ TOP DEVELOPMENTS [1] 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 [2] 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 [3] Chick-fil-A data breach affects more than 13,000 customers BleepingComputer · Incident · https://bleepingcomputer.com/news/security/chick-fil-a-data-breach-affects-more-than-13-000-customers SAST [4] Ethereum NFT Smart Contracts: Knowledge-Guided Vulnerability Detection with LLM and Code Slicing arXiv · Research · https://arxiv.org/abs/2607.21983 [5] KaPilot: LLM-Assisted Generation of Kani Specifications for Unsafe Rust Verification arXiv · Research · https://arxiv.org/abs/2607.21957 SCA [6] 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 [7] BlueNoroff Zoom Phishing Kit Profiles Crypto Wallets Before Malware Delivery The Hacker News · Analysis · https://thehackernews.com/2026/07/bluenoroff-zoom-phishing-kit-profiles.html AGENTIC SDLC [8] 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 [9] Loop Engineering: Version-Bound, Evidence-Gated Coordination for Production Coding Agents arXiv · Research · https://arxiv.org/abs/2611.18571 [10] "Go Home Copilot, You're Drunk": Understanding Developer Responses to Agent-Generated Code Review Comments arXiv · Research · https://arxiv.org/abs/2607.21997 TRUST [11] ReCon: A Resource-Constrained Benchmark for LLM-Based Cybersecurity Compliance Across Ingestion and Retrieval Pipelines arXiv · Research · https://arxiv.org/abs/2607.22885 MODELS [12] SIREN (Luring LLMs onto the Rocks): PAIR-Driven Preference Manipulation in Web-RAG Recommenders arXiv · Research · https://arxiv.org/abs/2607.21951 [13] Inference-Time Mitigation of Adversarial Political Bias in Large Language Models arXiv · Research · https://arxiv.org/abs/2608.14629 AI SECURITY [14] Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture arXiv · Research · https://arxiv.org/abs/2607.22445 [15] ResumeShield: Channel Separation and an Open Benchmark for Indirect Prompt Injection in AI Resume Screening arXiv · Research · https://arxiv.org/abs/2609.20188 [16] Agent Security Needs Redefinition through a Holistic Framework arXiv · Research · https://arxiv.org/abs/2607.22024 ADVISORIES IN THIS WINDOW 54 npm advisories · 12 critical · 27 high · 15 moderate 36 NuGet advisories · 2 critical · 7 high · 9 moderate · 18 low 34 PyPI advisories · 17 high · 12 moderate · 5 low 20 Go advisories · 1 critical · 6 high · 13 moderate 12 Maven advisories · 4 critical · 5 high · 3 moderate 8 crates.io advisories · 2 high · 5 moderate · 1 low 5 Packagist advisories · 1 critical · 4 high 2 Multi-ecosystem advisories · 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.