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 21 of 2026 · Retrospective · Covers 12 May 2026 11:00 to 19 May 2026 11:00 UTC

Read on the web

Top developments

Cisco warns of new critical SD-WAN flaw exploited in zero-day attacks

BleepingComputer · Incident · Remediation

New Windows 'MiniPlasma' zero-day exploit gives SYSTEM access, PoC released

BleepingComputer · Research · Remediation

Shai-Hulud: Here We Go Again. Mass npm Supply Chain Attack Hits the AntV Ecosystem

StepSecurity · Incident · SCA

Backdoored node-ipc npm releases steal developer credentials through DNS queries

Datadog · Incident · SCA

Grafana says stolen GitHub token let hackers steal codebase

BleepingComputer · Incident · SCA

SAST

Three Heads Are Better Than One: A Multi-perspective Reasoning Framework for Enhanced Vulnerability Detection

arXiv · Research · SAST

LLM-Based Static Verification of Code Against Natural-Language Requirements: An Industrial Experience Report

arXiv · Research · SAST

Speed Kills: Exploring Confused Deputy Attacks Through Edge AI Accelerators

arXiv · Research · SAST

Compile-time Security Analysis and Optimization of Sensitive String Producers

arXiv · Research · Agentic SDLC, SAST

SCA

actions-cool/issues-helper GitHub Action Compromised: All Tags Point to Imposter Commit That Exfiltrates CI/CD Credentials

StepSecurity · Incident · SCA

Active Supply Chain Attack: Malicious node-ipc Versions Published to npm

StepSecurity · Incident · SCA

Leaked Shai-Hulud malware fuels new npm infostealer campaign

BleepingComputer · Incident · SCA

Backdoored Cemu release linked to TanStack and Mistral supply chain campaign

Datadog · Incident · SCA

DAST

Stop Starving or Stuffing Me: Boosting Firmware Fuzzing Efficiency with On-demand Input Delivery

arXiv · Research · DAST

FuzzAgent: Multi-Agent System for End-to-End Evolutionary Library Fuzzing

arXiv · Research · DAST

Agentic SDLC

Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks

arXiv · Research · Agentic SDLC, AI Security

LivePI: More Realistic Benchmarking of Agents Against Indirect Prompt Injection

arXiv · Research · Agentic SDLC, AI Security

Rethinking Code Review in the Age of AI: A Vision for Agentic Code Review

arXiv · Research · Agentic SDLC

MemRepair: Hierarchical Memory for Agentic Repository-Level Vulnerability Repair

arXiv · Research · Agentic SDLC, Remediation

Remediation

Microsoft Patches 138 Vulnerabilities, Including DNS and Netlogon RCE Flaws

The Hacker News · Advisory · Remediation

New critical Exim mailer flaw allows remote code execution

BleepingComputer · Advisory · Remediation

Fragnesia: Linux Kernel Local Privilege Escalation via ESP-in-TCP

Wiz · Analysis · Remediation

Raising the bar: Quality, shared responsibility, and the future of GitHub’s bug bounty program

GitHub · Research · Remediation

Trust

Undermining the trust boundary: Investigating a stealthy intrusion through third-party compromise

Microsoft · Incident · Trust

Taking Stock of the State of European Cyber Resilience Act (CRA) Compliance: An Urgent Wake-up Call for the Open Source Ecosystem

OpenSSF · Standard / policy · Trust

5 Steps to Managing Shadow AI Tools Without Slowing Down Employees

BleepingComputer · Analysis · AI Security, Trust

From Obligation to Specification: A Survey on Validating EU AI Act Requirements in RE

arXiv · Research · Trust

Models

OriginBlame: Record- and Token-Level Data Provenance for AI Training Datasets

arXiv · Research · Models

Adaptive Probe-based Steering for Robust LLM Jailbreaking

arXiv · Research · Models

The Evaluation Game: Beyond Static LLM Benchmarking

arXiv · Research · Models

RoboJailBench: Benchmarking Adversarial Attacks and Defenses in Embodied Robotic Agents

arXiv · Research · Models

AI Security

On the Geometric Limits of Transformer Defenses against Obfuscation Attacks: Latent Embedding Collapse & Performance Robustness Gap

arXiv · Research · AI Security

Agent Security is a Systems Problem

arXiv · Research · AI Security, Models

Prompts Don't Protect: Architectural Enforcement via MCP Proxy for LLM Tool Access Control

arXiv · Research · AI Security

An Empirical Study of Privacy Leakage Chains via Prompt Injection in Black-Box Chatbot Environments

arXiv · Research · AI Security

Advisories in this window

  • 293 npm advisories · 208 critical · 50 high · 28 moderate · 7 low
  • 90 PyPI advisories · 14 critical · 40 high · 32 moderate · 4 low
  • 88 Go advisories · 7 critical · 34 high · 35 moderate · 12 low
  • 25 Maven advisories · 7 critical · 11 high · 6 moderate · 1 low
  • 22 Packagist advisories · 1 critical · 9 high · 10 moderate · 2 low
  • 18 NuGet advisories · 1 critical · 5 high · 12 moderate
  • 6 Hex advisories · 5 high · 1 moderate
  • 6 crates.io advisories · 1 critical · 4 high · 1 moderate
  • 5 Multi-ecosystem advisories · 3 critical · 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.