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 20 March 2026 Retrospective Covers 19 March 2026 00:00 to 20 March 2026 00:00 UTC Web: https://keryn.dev/headlines/daily/2026-03-20/ TOP DEVELOPMENTS [1] Critical Microsoft SharePoint flaw now exploited in attacks BleepingComputer · Incident · https://bleepingcomputer.com/news/microsoft/critical-microsoft-sharepoint-flaw-now-exploited-in-attacks [2] ThreatsDay Bulletin: FortiGate RaaS, Citrix Exploits, MCP Abuse, LiveChat Phish & More The Hacker News · Research · https://thehackernews.com/2026/03/threatsday-bulletin-fortigate-raas.html [3] A Novel Solution for Zero-Day Attack Detection in IDS using Self-Attention and Jensen-Shannon Divergence in WGAN-GP arXiv · Research · https://arxiv.org/abs/2603.19350 SCA [4] When AI Agrees Too Much: Manipulating Automated Code Review via Bias Injection arXiv · Research · https://arxiv.org/abs/2603.18740 DAST [5] Weaver: Fuzzing JavaScript Engines at the JavaScript-WebAssembly Boundary arXiv · Research · https://arxiv.org/abs/2603.18789 [6] TENSURE: Fuzzing Sparse Tensor Compilers (Registered Report) arXiv · Research · https://arxiv.org/abs/2603.18372 AGENTIC SDLC [7] How Squad runs coordinated AI agents inside your repository GitHub · Analysis · https://github.blog/ai-and-ml/github-copilot/how-squad-runs-coordinated-ai-agents-inside-your-repository [8] How Ceros Gives Security Teams Visibility and Control in Claude Code The Hacker News · Analysis · https://thehackernews.com/2026/03/how-ceros-gives-security-teams.html REMEDIATION [9] Max severity Ubiquiti UniFi flaw may allow account takeover BleepingComputer · Analysis · https://bleepingcomputer.com/news/security/ubiquiti-warns-of-unifi-flaw-that-may-enable-account-takeover [10] Cross-Ecosystem Vulnerability Analysis for Python Applications arXiv · Research · https://arxiv.org/abs/2603.18693 TRUST [11] New tools and guidance: Announcing Zero Trust for AI Microsoft · Standard / policy · https://microsoft.com/en-us/security/blog/2026/03/19/new-tools-and-guidance-announcing-zero-trust-for-ai [12] The Causal Impact of Tool Affordance on Safety Alignment in LLM Agents arXiv · Research · https://arxiv.org/abs/2603.20320 [13] Security awareness in LLM agents: the NDAI zone case arXiv · Research · https://arxiv.org/abs/2603.19011 MODELS [14] The Autonomy Tax: Defense Training Breaks LLM Agents arXiv · Research · https://arxiv.org/abs/2603.19423 [15] Automated Membership Inference Attacks (AutoMIA): Discovering MIA Signal Computations using LLM Agents arXiv · Research · https://arxiv.org/abs/2603.19375 [16] On Optimizing Multimodal Jailbreaks for Spoken Language Models arXiv · Research · https://arxiv.org/abs/2603.19127 AI SECURITY [17] A Framework for Formalizing LLM Agent Security arXiv · Research · https://arxiv.org/abs/2603.19469 [18] Prompt Control-Flow Integrity: A Priority-Aware Runtime Defense Against Prompt Injection in LLM Systems arXiv · Research · https://arxiv.org/abs/2603.18433 ADVISORIES IN THIS WINDOW 17 Go advisories · 3 critical · 7 high · 7 moderate 16 npm advisories · 9 critical · 3 high · 4 moderate 16 Packagist advisories · 2 critical · 5 high · 8 moderate · 1 low 11 PyPI advisories · 1 critical · 6 high · 4 moderate 3 NuGet advisories · 2 high · 1 moderate 2 RubyGems advisories · 1 high · 1 moderate 2 crates.io advisories · 2 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.