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 19 of 2025 · Retrospective · Covers 29 April 2025 11:00 to 6 May 2025 11:00 UTC

Read on the web

Top developments

Hundreds of e-commerce sites hacked in supply-chain attack

Ars Technica: Security · Incident · SCA

Linux wiper malware hidden in malicious Go modules on GitHub

BleepingComputer · Analysis · SCA

Magento supply chain attack compromises hundreds of e-stores

BleepingComputer · Incident · SCA

Malicious PyPI packages abuse Gmail, websockets to hijack systems

BleepingComputer · Incident · SCA

Wormable AirPlay Flaws Enable Zero-Click RCE on Apple Devices via Public Wi-Fi

The Hacker News · Research · Remediation

SAST

Poster: Machine Learning for Vulnerability Detection as Target Oracle in Automated Fuzz Driver Generation

arXiv · Research · DAST, SAST

Security-by-Design at the Telco Edge with OSS: Challenges and Lessons Learned

arXiv · Research · Remediation, SAST

LASHED: LLMs And Static Hardware Analysis for Early Detection of RTL Bugs

arXiv · Research · SAST

SCA

Causes and Canonicalization of Unreproducible Builds in Java

arXiv · Research · SCA

Unveiling Large Language Model Supply Chain: Structure, Domain, and Vulnerabilities

arXiv · Research · SCA

AI-generated code could be a disaster for the software supply chain. Here’s why.

Ars Technica: Security · Analysis · SCA

Native Support For SBOMs In Python (PEP-770) Is A Reality — But Not Risk-Free

Checkmarx · Analysis · SCA

DAST

Testing SSD Firmware with State Data-Aware Fuzzing: Accelerating Coverage in Nondeterministic I/O Environments

arXiv · Research · DAST

Targeted Fuzzing for Unsafe Rust Code: Leveraging Selective Instrumentation

arXiv · Research · DAST

Refining Fuzzed Crashing Inputs for Better Fault Diagnosis

arXiv · Research · DAST

Why MFA is getting easier to bypass and what to do about it

Ars Technica: Security · Analysis · DAST

Agentic SDLC

SecRepoBench: Benchmarking Code Agents for Secure Code Completion in Real-World Repositories

arXiv · Research · Agentic SDLC

The AI-powered DevOps revolution: Redefining developer collaboration

GitHub · Analysis · Agentic SDLC

Product Walkthrough: Securing Microsoft Copilot with Reco

The Hacker News · Analysis · Agentic SDLC

Skyhigh Security adds data protection solutions for Microsoft Copilot and ChatGPT Enterprise

Help Net Security · Analysis · Agentic SDLC

Remediation

Microsoft pushes fix for Windows 11 24H2 update failures

BleepingComputer · Analysis · Remediation

BadPatches: Routing-Aware Backdoor Attacks on Vision Mixture-of-Experts

arXiv · Research · Remediation

PatchFuzz: Patch Fuzzing for JavaScript Engines

arXiv · Research · Remediation

DARPA believes AI Cyber Challenge could upend patching as the industry knows it

CyberScoop · Analysis · Remediation

Trust

Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents

arXiv · Research · AI Security, Models, Trust

Federal Data, Meet your New Bodyguard: DSPM joins Wiz for Government

Wiz · Analysis · Trust

Verifying Sigstore Bundles as an End User

Sigstore · Analysis · Trust

Snyk Joins Coalition for Secure AI (CoSAI) to Champion Safe, Responsible AI Development

CoSAI · Analysis · Trust

Models

A Chaos Driven Metric for Backdoor Attack Detection

arXiv · Research · Models

AKD : Adversarial Knowledge Distillation For Large Language Models Alignment on Coding tasks

arXiv · Research · Models

A Survey on Privacy Risks and Protection in Large Language Models

arXiv · Research · Models

Adversarial Robustness of Deep Learning Models for Inland Water Body Segmentation from SAR Images

arXiv · Research · Models

AI Security

A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)

arXiv · Research · AI Security

MCP: Untrusted Servers and Confused Clients, Plus a Sneaky Exploit

Embrace The Red · Analysis · AI Security

OET: Optimization-based prompt injection Evaluation Toolkit

arXiv · Research · AI Security, Models

The Illusion of Role Separation: Hidden Shortcuts in LLM Role Learning (and How to Fix Them)

arXiv · Research · AI Security

Advisories in this window

  • 131 npm advisories · 122 critical · 1 high · 5 moderate · 3 low
  • 13 Maven advisories · 5 critical · 2 high · 4 moderate · 2 low
  • 13 Packagist advisories · 3 critical · 3 high · 4 moderate · 3 low
  • 9 Go advisories · 3 high · 6 moderate
  • 5 PyPI advisories · 1 critical · 1 high · 3 moderate
  • 3 crates.io advisories · 1 high · 1 moderate · 1 low

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.