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

Read on the web

Top developments

Critical Microsoft SharePoint flaw now exploited in attacks

BleepingComputer · Incident · Remediation

ThreatsDay Bulletin: FortiGate RaaS, Citrix Exploits, MCP Abuse, LiveChat Phish & More

The Hacker News · Research · AI Security

A Novel Solution for Zero-Day Attack Detection in IDS using Self-Attention and Jensen-Shannon Divergence in WGAN-GP

arXiv · Research · Remediation

SCA

When AI Agrees Too Much: Manipulating Automated Code Review via Bias Injection

arXiv · Research · SCA

DAST

Weaver: Fuzzing JavaScript Engines at the JavaScript-WebAssembly Boundary

arXiv · Research · DAST

TENSURE: Fuzzing Sparse Tensor Compilers (Registered Report)

arXiv · Research · DAST

Agentic SDLC

How Squad runs coordinated AI agents inside your repository

GitHub · Analysis · Agentic SDLC

How Ceros Gives Security Teams Visibility and Control in Claude Code

The Hacker News · Analysis · Agentic SDLC

Remediation

Max severity Ubiquiti UniFi flaw may allow account takeover

BleepingComputer · Analysis · Remediation

Cross-Ecosystem Vulnerability Analysis for Python Applications

arXiv · Research · Remediation

Trust

New tools and guidance: Announcing Zero Trust for AI

Microsoft · Standard / policy · Trust

The Causal Impact of Tool Affordance on Safety Alignment in LLM Agents

arXiv · Research · Models, Trust

Security awareness in LLM agents: the NDAI zone case

arXiv · Research · Trust

Models

The Autonomy Tax: Defense Training Breaks LLM Agents

arXiv · Research · AI Security, Models

Automated Membership Inference Attacks (AutoMIA): Discovering MIA Signal Computations using LLM Agents

arXiv · Research · Models

On Optimizing Multimodal Jailbreaks for Spoken Language Models

arXiv · Research · Models

AI Security

A Framework for Formalizing LLM Agent Security

arXiv · Research · AI Security

Prompt Control-Flow Integrity: A Priority-Aware Runtime Defense Against Prompt Injection in LLM Systems

arXiv · Research · AI Security, Models

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

Titles, links and short excerpts only; every item links to its source. Selection is automatic; the editor's note is AI-drafted and checked as described. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.