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 22 of 2025 · Retrospective · Covers 20 May 2025 11:00 to 27 May 2025 11:00 UTC

Read on the web

Top developments

Chinese hackers breach US local governments using Cityworks zero-day

BleepingComputer · Incident · Remediation

GitLab Duo Vulnerability Enabled Attackers to Hijack AI Responses with Hidden Prompts

The Hacker News · Incident · Agentic SDLC, AI Security

GitHub MCP Exploited: Accessing private repositories via MCP

Simon Willison · Incident · AI Security

Dozens of malicious packages on NPM collect host and network data

BleepingComputer · Analysis · SCA

Trojanized RVTools push Bumblebee malware in SEO poisoning campaign

BleepingComputer · Incident · SCA

SAST

Eradicating the Unseen: Detecting, Exploiting, and Remediating a Path Traversal Vulnerability across GitHub

arXiv · Research · Remediation, SAST, SCA

Exposing Go's Hidden Bugs: A Novel Concolic Framework

arXiv · Research · DAST, SAST

An Empirical Study of JavaScript Inclusion Security Issues in Chrome Extensions

arXiv · Research · SAST

Learning to Focus: Context Extraction for Efficient Code Vulnerability Detection with Language Models

arXiv · Research · SAST

SCA

Data-stealing Chrome extensions impersonate Fortinet, YouTube, VPNs

BleepingComputer · Analysis · SCA

Destructive malware available in NPM repo went unnoticed for 2 years

Ars Technica: Security · Research · SCA

ARMS: A Vision for Actor Reputation Metric Systems in the Open-Source Software Supply Chain

arXiv · Research · SCA

Software Security Mapping Framework: Operationalization of Security Requirements

arXiv · Research · SCA, Trust

DAST

Penetration Testing for System Security: Methods and Practical Approaches

arXiv · Research · DAST

The Crowded Battle: Key Insights from the 2025 State of Pentesting Report

The Hacker News · Analysis · DAST

Agentic SDLC

Taming System Complexity: Demystifying Software Engineering Agents in Diagnosing Linux Kernel Faults

arXiv · Research · Agentic SDLC

Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AI

arXiv · Research · Agentic SDLC

Tool Preferences in Agentic LLMs are Unreliable

arXiv · Research · Agentic SDLC

Agent mode 101: All about GitHub Copilot’s powerful mode

GitHub · Analysis · Agentic SDLC

Remediation

Ivanti EPMM RCE Vulnerability Chain Exploited in the Wild

Wiz · Incident · Remediation

Bypassing MTE with CVE-2025-0072

GitHub · Analysis · Remediation

LLM-BSCVM: An LLM-Based Blockchain Smart Contract Vulnerability Management Framework

arXiv · Research · Remediation, SAST

VIVID: A Novel Approach to Remediation Prioritization in Static Application Security Testing (SAST)

arXiv · Research · Remediation, SAST

Trust

Engineering Trustworthy Machine-Learning Operations with Zero-Knowledge Proofs

arXiv · Research · Trust

A Systematic Classification of Vulnerabilities in MoveEVM Smart Contracts (MWC)

arXiv · Research · Trust

AWS re:Inforce Sessions Highlight Secure AI and Industry Collaboration

CoSAI · Analysis · Trust

Deloitte’s Secure by Design (SbD) Approach – Enhanced with Wiz

Wiz · Analysis · Trust

Models

MultiPhishGuard: An Explainable and Adaptive Multi-Agent LLM System for Phishing Email Detection

arXiv · Research · Models

Lifelong Safety Alignment for Language Models

arXiv · Research · Models

Capability-Based Scaling Trends for LLM-Based Red-Teaming

arXiv · Research · AI Security, Models

Poison in the Well: Feature Embedding Disruption in Backdoor Attacks

arXiv · Research · Models

AI Security

Researchers cause GitLab AI developer assistant to turn safe code malicious

Ars Technica: Security · Research · AI Security

The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent Homology

arXiv · Research · AI Security

Stronger Enforcement of Instruction Hierarchy via Augmented Intermediate Representations

arXiv · Research · AI Security

A Critical Evaluation of Defenses against Prompt Injection Attacks

arXiv · Research · AI Security

Advisories in this window

  • 384 npm advisories · 383 critical · 1 moderate
  • 14 Packagist advisories · 1 critical · 4 high · 5 moderate · 4 low
  • 10 Go advisories · 3 high · 5 moderate · 2 low
  • 7 PyPI advisories · 1 critical · 1 high · 5 moderate
  • 4 crates.io advisories · 4 low
  • 3 NuGet advisories · 2 moderate · 1 low
  • 3 Maven advisories · 1 critical · 2 moderate
  • 1 Multi-ecosystem advisory · 1 moderate

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.