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 51 of 2025 · Retrospective · Covers 9 December 2025 12:00 to 16 December 2025 12:00 UTC

Read on the web

Top developments

Google fixes eighth Chrome zero-day exploited in attacks in 2025

BleepingComputer · Incident · Remediation

CISA orders feds to patch actively exploited Geoserver flaw

BleepingComputer · Advisory · Remediation

Microsoft December 2025 Patch Tuesday fixes 3 zero-days, 57 flaws

BleepingComputer · Advisory · Remediation

Apple fixes two zero-day flaws exploited in 'sophisticated' attacks

BleepingComputer · Incident · Remediation

Shai-Hulud 2.0: Guidance for detecting, investigating, and defending against the supply chain attack

Microsoft · Incident · SCA

SAST

From Obfuscated to Obvious: A Comprehensive JavaScript Deobfuscation Tool for Security Analysis

arXiv · Research · SAST

SHERLOCK: A Deep Learning Approach To Detect Software Vulnerabilities

arXiv · Research · SAST

Taint-Based Code Slicing for LLMs-based Malicious NPM Package Detection

arXiv · Research · SAST, SCA

Llama-based source code vulnerability detection: Prompt engineering vs Fine tuning

arXiv · Research · SAST

SCA

Inside Shai-Hulud’s Maw: How The NPM Worm Exploits And Propagates

Checkmarx · Incident · SCA

Unveiling Malicious Logic: Towards a Statement-Level Taxonomy and Dataset for Securing Python Packages

arXiv · Research · SCA

Cargo Sherlock: An SMT-Based Checker for Software Trust Costs

arXiv · Research · SCA

Keep the Lights On, Keep the Lengths in Check: Plug-In Adversarial Detection for Time-Series LLMs in Energy Forecasting

arXiv · Research · Models, Remediation, SCA

DAST

PentestEval: Benchmarking LLM-based Penetration Testing with Modular and Stage-Level Design

arXiv · Research · DAST

The Role of AI in Modern Penetration Testing

arXiv · Research · DAST

Automated Penetration Testing with LLM Agents and Classical Planning

arXiv · Research · Agentic SDLC, DAST

Comparing AI Agents to Cybersecurity Professionals in Real-World Penetration Testing

arXiv · Research · Agentic SDLC, AI Security, DAST

Agentic SDLC

Re-opening open-source science through AI assisted development

arXiv · Research · Agentic SDLC

Remediation

Google and Apple roll out emergency security updates after zero-day attacks

TechCrunch: Security · Analysis · Remediation

Chrome Targeted by Active In-the-Wild Exploit Tied to Undisclosed High-Severity Flaw

The Hacker News · Analysis · Remediation

Microsoft Issues Security Fixes for 56 Flaws, Including Active Exploit and Two Zero-Days

The Hacker News · Analysis · Remediation

Notepad++ fixes flaw that let attackers push malicious update files

BleepingComputer · Research · Remediation

Trust

Beyond Procedural Compliance: Human Oversight as a Dimension of Well-being Efficacy in AI Governance

arXiv · Research · Trust

From Linear Risk to Emergent Harm: Complexity as the Missing Core of AI Governance

arXiv · Research · Trust

AI Transparency Atlas: Framework, Scoring, and Real-Time Model Card Evaluation Pipeline

arXiv · Research · Models, Trust

Verification of Lightning Network Channel Balances with Trusted Execution Environments (TEE)

arXiv · Research · Trust

Models

CIS-BA: Continuous Interaction Space Based Backdoor Attack for Object Detection in the Real-World

arXiv · Research · Models

Bilevel Optimization for Covert Memory Tampering in Heterogeneous Multi-Agent Architectures (XAMT)

arXiv · Research · Models

The Double Life of Code World Models: Provably Unmasking Malicious Behavior Through Execution Traces

arXiv · Research · Models

Comparative Analysis of LLM Abliteration Methods: A Cross-Architecture Evaluation

arXiv · Research · Models

AI Security

IntentMiner: Intent Inversion Attack via Tool Call Analysis in the Model Context Protocol

arXiv · Research · AI Security

Cisco Integrated AI Security and Safety Framework Report

arXiv · Research · AI Security

ceLLMate: Sandboxing Browser AI Agents

arXiv · Research · AI Security

One Leak Away: How Pretrained Model Exposure Amplifies Jailbreak Risks in Finetuned LLMs

arXiv · Research · AI Security, Models

Advisories in this window

  • 134 npm advisories · 115 critical · 9 high · 8 moderate · 2 low
  • 29 Maven advisories · 2 critical · 12 high · 13 moderate · 2 low
  • 21 Go advisories · 1 critical · 11 high · 9 moderate
  • 18 PyPI advisories · 2 critical · 5 high · 8 moderate · 3 low
  • 11 Packagist advisories · 5 critical · 4 high · 1 moderate · 1 low
  • 2 crates.io advisories · 2 moderate
  • 2 NuGet advisories · 1 high · 1 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.