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 17 July 2026 Retrospective Covers 16 July 2026 00:00 to 17 July 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-07-17/ TOP DEVELOPMENTS [1] CISA orders feds to patch actively exploited Oracle flaw by Saturday BleepingComputer · Advisory · https://bleepingcomputer.com/news/security/cisa-orders-feds-to-patch-actively-exploited-oracle-flaw-by-saturday [2] Unpacking the AsyncAPI npm supply chain compromise and import-time payload delivery Microsoft · Incident · https://microsoft.com/en-us/security/blog/2026/07/15/unpacking-asyncapi-npm-supply-chain-compromise-import-time-payload-delivery [3] CVE-2026-39808: Fortinet FortiSandbox OS Command Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog CISA · Advisory · https://cisa.gov/known-exploited-vulnerabilities-catalog?search_api_fulltext=CVE-2026-39808 SCA [4] Setup Complete, Now You Are Compromised: Weaponizing Setup Instructions Against AI Coding Agents arXiv · Research · https://arxiv.org/abs/2607.15143 [5] The Distributed Open-Source Vulnerability Ecosystem arXiv · Research · https://arxiv.org/abs/2607.14900 DAST [6] Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents arXiv · Research · https://arxiv.org/abs/2607.15263 [7] AI Can Find Bugs, But Human Knowledge Still Proves Them The Hacker News · Analysis · https://thehackernews.com/2026/07/ai-can-find-bugs-but-human-knowledge.html AGENTIC SDLC [8] Bad Memory: Evaluating Prompt Injection Risks from Memory in Agentic Systems arXiv · Research · https://arxiv.org/abs/2607.14611 [9] Democratizing Agent Deployment Safety: A Structural Monitoring Approach arXiv · Research · https://arxiv.org/abs/2607.14570 [10] New Agent Data Injection Attack Can Make AI Agents Misclick or Run Attacker Commands The Hacker News · Analysis · https://thehackernews.com/2026/07/new-agent-data-injection-attack-can.html TRUST [11] SCITUS: A Multi-Jurisdictional Framework for Adapting NIST AI RMF to the Canadian Regulatory Context arXiv · Research · https://arxiv.org/abs/2607.15051 [12] StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows arXiv · Research · https://arxiv.org/abs/2607.14896 [13] Global Index on Responsible AI 2026 : Conceptual Framework and Methodology arXiv · Research · https://arxiv.org/abs/2608.18122 MODELS [14] When Words Are Safe But Actions Kill: Probing Physical Jailbreak Beyond Textual Jailbreak in Hidden-State Risk Space arXiv · Research · https://arxiv.org/abs/2607.15218 AI SECURITY [15] FlowGuard: From Signals to Evidence for MCP Security Detection arXiv · Research · https://arxiv.org/abs/2607.14754 [16] From Indirect Prompt Injection to DNS Exfiltration in macOS Terminal Embrace The Red · Analysis · https://embracethered.com/blog/posts/2026/macos-terminal-dillma-dns-exfil-ansi-escape-code-fix [17] Context Contamination in LLM Analysis of Network Security Logs: Poison with Passive Prompt Injection and Mitigation Evaluation arXiv · Research · https://arxiv.org/abs/2607.14493 ADVISORIES IN THIS WINDOW 23 npm advisories · 22 critical · 1 high 11 Go advisories · 1 critical · 2 high · 8 moderate 5 PyPI advisories · 2 critical · 3 high 4 Maven advisories · 4 high 3 Packagist advisories · 1 critical · 2 high 3 crates.io advisories · 1 moderate · 2 low 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.