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, Saturday 9 May 2026 · Retrospective · Covers 8 May 2026 00:00 to 9 May 2026 00:00 UTC

Read on the web

Top developments

Flaw in Claude’s Chrome extension allowed ‘any’ other plugin to hijack victims’ AI

CyberScoop · Incident · Agentic SDLC

Detecting Web Server Probing & Fuzzing in Traefik with Automated Cloudflare Response

Elastic · Research · DAST

Quasar Linux RAT Steals Developer Credentials for Software Supply Chain Compromise

The Hacker News · Incident · SCA

SAST

Longitudinal Analyses of SAST Tools: A CodeQL Case Study

arXiv · Research · SAST

Unsafe by Flow: Uncovering Bidirectional Data-Flow Risks in MCP Ecosystem

arXiv · Research · AI Security, SAST

Securing the Dark Matter: A Semantic-Enhanced Neuro-Symbolic Framework for Supply Chain Analysis of Opaque Industrial Software

arXiv · Research · SAST

SCA

Kettle: Attested builds for verifiable software provenance

arXiv · Research · SCA

Seed Hijacking of LLM Sampling and Quantum Random Number Defense

arXiv · Research · SCA

Cross-Modal Backdoors in Multimodal Large Language Models

arXiv · Research · Models, SCA

Agentic SDLC

Using Claude Code: The Unreasonable Effectiveness of HTML

Simon Willison · Analysis · Agentic SDLC

SecureForge: Finding and Preventing Vulnerabilities in LLM-Generated Code via Prompt Optimization

arXiv · Research · Agentic SDLC

Collaborator or Assistant? How AI Coding Agents Partition Work Across Pull Request Lifecycles

arXiv · Research · Agentic SDLC, Trust

Remediation

Demystifying and Detecting Agentic Workflow Injection Vulnerabilities in GitHub Actions

arXiv · Research · Remediation, SCA

Models

When Child Inherits: Modeling and Exploiting Subagent Spawn in Multi-Agent Networks

arXiv · Research · AI Security, Models

The Attacker in the Mirror: Breaking Self-Consistency in Safety via Anchored Bipolicy Self-Play

arXiv · Research · Models

GLiGuard: Schema-Conditioned Classification for LLM Safeguard

arXiv · Research · Models

AI Security

LLM Advertisement based on Neuron Auctions

arXiv · Research · AI Security

WebTrap: Stealthy Mid-Task Hijacking of Browser Agents During Navigation

arXiv · Research · AI Security

Advisories in this window

  • 36 PyPI advisories · 2 critical · 12 high · 21 moderate · 1 low
  • 28 Go advisories · 6 critical · 8 high · 12 moderate · 2 low
  • 21 npm advisories · 5 critical · 10 high · 6 moderate
  • 8 Packagist advisories · 2 critical · 2 high · 4 moderate
  • 7 Maven advisories · 4 high · 3 moderate
  • 3 RubyGems advisories · 3 moderate
  • 3 Hex advisories · 2 high · 1 low
  • 2 NuGet advisories · 2 moderate
  • 2 crates.io advisories · 1 critical · 1 high
  • 1 Multi-ecosystem 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.