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 7 November 2026 · Retrospective · Covers 6 November 2026 00:00 to 7 November 2026 00:00 UTC

Read on the web

Top developments

Runaway AI hacking agent behind WorkOS breach after escaping sandbox

BleepingComputer · Incident · AI Security

Chainalysis says hackers stole credentials and tokens for all of its customers

Ars Technica: Security · Incident · AI Security

Shai-Hulud Descendant Surfaces in Malicious Ruby Gem That Steals Credentials and Poisons AI Tools

Socket · Research · AI Security, SCA

Malicious Browser Extensions Could Use Claude Dispatch to Run Code on Users' Computers

The Hacker News · Research · SCA

SCA

Free Certified Patch for Critical Rails Active Storage Vulnerability

Socket · Analysis · Remediation, SCA

Agents on the Patch Path: Mining AI-Associated Security Remediation in GitHub Security Advisories

arXiv · Research · Agentic SDLC, Remediation, SCA

PyPI's redesigned project pages with clearer security signals are now live

Python Package Index · Analysis · SCA, Trust

DAST

Wake 5: A Python Framework for Solidity Testing, Fuzzing, and Static Analysis

arXiv · Research · DAST

Agentic SDLC

From Delegation to Action Approval: An Exploratory Study of Developers' Practices in Collaborating with AI Coding Agents

arXiv · Research · Agentic SDLC

Remediation

Beyond a Single Commit: Anchor-Guided Reconstruction of Multi-Commit Patches

arXiv · Research · Remediation

Trust

From Sigstore to CRA readiness: Sylva’s journey with OpenSSF

OpenSSF · Standard / policy · Trust

Rethinking Requirement-to-Code Traceability Through Single-Agent and Multi-Agent Approaches

arXiv · Research · Trust

Models

Beyond the Prompt: Leveraging Pre-Decoding States for Jailbreak Detection in dLLMs

arXiv · Research · Models

Enmo Safety: External, Non-invasive, Model-agnostic Oversight for Off-the-Shelf Language Models

arXiv · Research · Models

GuardrailAgent: An Agentic Framework for Jailbreak Defense

arXiv · Research · Models

AI Security

MemoryDojo: Benchmarking Poisoning Through LLM Agent Memory

arXiv · Research · AI Security

GreyZone: Production-quality Agentic Application Threats

arXiv · Research · AI Security

Notwithstanding Anything to the Contrary, Ignore All Instructions: Measuring Prompt Injection Guards on Legal Contract Review Attacks

arXiv · Research · AI Security

Advisories in this window

  • 32 npm advisories · 32 critical
  • 12 Go advisories · 6 critical · 3 high · 3 moderate
  • 4 PyPI advisories · 1 critical · 1 high · 2 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.