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

Daily, Wednesday 3 June 2026 · Retrospective · Covers 2 June 2026 00:00 to 3 June 2026 00:00 UTC

Read on the web

Top developments

Google fixes one actively exploited Android zero-day, 124 flaws

BleepingComputer · Incident · Remediation

CISA flags two-year-old Oracle flaw as actively exploited in attacks

BleepingComputer · Incident · Remediation

Nx Console VS Code Extension Compromised

StepSecurity · Incident · SCA

Multiple redhat-cloud-services npm Packages compromised

StepSecurity · Incident · SCA

Laravel-Lang Supply Chain Attack: Every Tag Across Multiple Composer Packages Rewritten to Steal CI Secrets

StepSecurity · Incident · SCA

SCA

Dev Machine Guard Now Scans Extensions Across Every Modern IDE

StepSecurity · Analysis · Agentic SDLC, SCA

RogueMerge: Robust and Unified Attacks against LLM Model Merging

arXiv · Research · Models, SCA

The case for GitHub Actions security after recent supply chain attacks

Datadog · Analysis · SCA

Agentic SDLC

The Impact of Configuring Agentic AI Coding Tools on Build-vs-Buy Decisions: A Study Protocol

arXiv · Research · Agentic SDLC

Remediation

AI Agents Enable Adaptive Computer Worms

arXiv · Research · Remediation

FORGE: Multi-Agent Graduated Exploitation and Detection Engineering

arXiv · Research · Remediation

Patcher: Post-Hoc Patching of Backdoored Large Language Models

arXiv · Research · Models, Remediation

Trust

Proof-Carrying Agent Actions: Model-Agnostic Runtime Governance for Heterogeneous Agent Systems

arXiv · Research · Trust

Why the browser is now the front line for AI security

BleepingComputer · Analysis · AI Security, Trust

Overlaying Governance: A Compositional Authorization Framework for Delegation and Scope in Agentic AI

arXiv · Research · Trust

Models

From Control Boundary to Insurance Claim: Reconstructing AI-Mediated Losses Through the CER Framework

arXiv · Research · AI Security, Models

Bayesian Membership Privacy for Graph Neural Networks

arXiv · Research · Models

All you need to break LLMs are Black-Box, Adapting, Efficient, Transferable, Harmful, Applicable ... Attacks

arXiv · Research · Models

AI Security

Notarized Agents: Receiver-Attested Confidential Receipts for AI Agent Actions

arXiv · Research · AI Security

Caught in the Act(ivation): Toward Pre-Output and Multi-Turn Detection of Credential Exfiltration by LLM Agents

arXiv · Research · AI Security

SkillHarness: Harnessing Safe Skills for Computer-Use Agents

arXiv · Research · AI Security

Advisories in this window

  • 6 PyPI advisories · 1 critical · 1 high · 2 moderate · 2 low
  • 2 npm advisories · 1 critical · 1 moderate
  • 1 Maven 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.