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 10 of 2026 · Retrospective · Covers 24 February 2026 12:00 to 3 March 2026 12:00 UTC

Read on the web

Top developments

Google addresses actively exploited Qualcomm zero-day in fresh batch of 129 Android vulnerabilities

CyberScoop · Incident · Remediation

ClawJacked attack let malicious websites hijack OpenClaw to steal data

BleepingComputer · Incident · AI Security

Critical SolarWinds Serv-U flaws offer root access to servers

BleepingComputer · Analysis · Remediation

CyberStrikeAI tool adopted by hackers for AI-powered attacks

BleepingComputer · Research · AI Security

Case Study: Defending the Open Source Supply Chain in a New Regulatory Era

OpenSSF · Research · Trust

SAST

Benchmark of Benchmarks: Unpacking Influence and Code Repository Quality in LLM Safety Benchmarks

arXiv · Research · AI Security, Models, SAST

ZeroDayBench: Evaluating LLM Agents on Unseen Zero-Day Vulnerabilities for Cyberdefense

arXiv · Research · Agentic SDLC, Remediation, SAST

Enhancing Continual Learning for Software Vulnerability Prediction: Addressing Catastrophic Forgetting via Hybrid-Confidence-Aware Selective Replay for Temporal LLM Fine-Tuning

arXiv · Research · SAST

Automated Vulnerability Detection in Source Code Using Deep Representation Learning

arXiv · Research · SAST

SCA

Malicious StripeApi NuGet Package Mimicked Official Library and Stole API Tokens

The Hacker News · Incident · SCA

Malicious NuGet Packages Stole ASP.NET Data; npm Package Dropped Malware

The Hacker News · Incident · SCA

Human-Certified Module Repositories for the AI Age

arXiv · Research · SCA

North Korean Hackers Publish 26 npm Packages Hiding Pastebin C2 for Cross-Platform RAT

The Hacker News · Research · SCA

DAST

NeuroSCA: Neuro-Symbolic Constraint Abstraction for Smart Contract Hybrid Fuzzing

arXiv · Research · DAST

AWE: Adaptive Agents for Dynamic Web Penetration Testing

arXiv · Research · DAST

Towards the Systematic Testing of Regular Expression Engines

arXiv · Research · DAST

LLM-Powered Silent Bug Fuzzing in Deep Learning Libraries via Versatile and Controlled Bug Transfer

arXiv · Research · DAST

Agentic SDLC

RoguePilot Flaw in GitHub Codespaces Enabled Copilot to Leak GITHUB_TOKEN

The Hacker News · Incident · Agentic SDLC, AI Security

A Systematic Study of LLM-Based Architectures for Automated Patching

arXiv · Research · Agentic SDLC, Remediation

ClarEval: A Benchmark for Evaluating Clarification Skills of Code Agents under Ambiguous Instructions

arXiv · Research · Agentic SDLC

Claude Code Flaws Allow Remote Code Execution and API Key Exfiltration

The Hacker News · Research · Agentic SDLC, AI Security

Remediation

Taming Agentic Browsers: Vulnerability in Chrome Allowed Extensions to Hijack New Gemini Panel

Palo Alto Networks · Incident · Remediation

Third-Party Patching and the Business Footprint We All Share

BleepingComputer · Analysis · Remediation

Formal Analysis and Supply Chain Security for Agentic AI Skills

arXiv · Research · Remediation

Trend Micro warns of critical Apex One code execution flaws

BleepingComputer · Analysis · Remediation

Trust

Beyond Detection: Governing GenAI in Academic Peer Review as a Sociotechnical Challenge

arXiv · Research · Trust

Lifecycle-Integrated Security for AI-Cloud Convergence in Cyber-Physical Infrastructure

arXiv · Research · Trust

Agent Behavioral Contracts: Formal Specification and Runtime Enforcement for Reliable Autonomous AI Agents

arXiv · Research · Trust

AWS successfully completed its first surveillance audit for ISO 42001:2023 with no findings

AWS Security Blog · Standard / policy · Trust

Models

Kraken: Higher-order EM Side-Channel Attacks on DNNs in Near and Far Field

arXiv · Research · Models

DSBA: Dynamic Stealthy Backdoor Attack with Collaborative Optimization in Self-Supervised Learning

arXiv · Research · Models

From Shallow to Deep: Pinning Semantic Intent via Causal GRPO

arXiv · Research · Models

SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety

arXiv · Research · Models

AI Security

Fooling AI Agents: Web-Based Indirect Prompt Injection Observed in the Wild

Palo Alto Networks · Research · AI Security

Extending the Formalism and Theoretical Foundations of Cryptography to AI

arXiv · Research · AI Security, Models

DualSentinel: A Lightweight Framework for Detecting Targeted Attacks in Black-box LLM via Dual Entropy Lull Pattern

arXiv · Research · AI Security

Tracking Capabilities for Safer Agents

arXiv · Research · AI Security

Advisories in this window

  • 285 npm advisories · 196 critical · 38 high · 40 moderate · 11 low
  • 52 Go advisories · 4 critical · 26 high · 19 moderate · 3 low
  • 52 PyPI advisories · 4 critical · 14 high · 30 moderate · 4 low
  • 37 NuGet advisories · 8 high · 23 moderate · 6 low
  • 17 Packagist advisories · 4 critical · 7 high · 5 moderate · 1 low
  • 10 crates.io advisories · 2 critical · 3 high · 5 moderate
  • 10 Maven advisories · 2 high · 4 moderate · 4 low
  • 1 Hex advisory · 1 low
  • 1 RubyGems advisory · 1 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.