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, Thursday 1 October 2026 · Retrospective · Covers 30 September 2026 00:00 to 1 October 2026 00:00 UTC

Read on the web

Top developments

Cisco warns of new SD-WAN zero-day exploited in attacks

BleepingComputer · Incident · Remediation

Sckit Supply Chain Worm Hits MemTensor npm & PyPi scopes

StepSecurity · Incident · SCA

TeamViewer urges users to patch severe flaws “as soon as possible”

BleepingComputer · Advisory · Remediation

Disrupting a coordinated model-distillation campaign

OpenAI News · Incident · Models

OpenSSL Fixes High-Severity DTLS Flaw That Can Leak Heap Memory Unencrypted

The Hacker News · Incident · Remediation

SAST

Pretext: Defeating Malicious Skill Detection Frameworks for AI Agents

arXiv · Research · Agentic SDLC, AI Security, SAST

SCA

Hiding in Plain Sight: Decoupling Pretext from Actuation for Skill Poisoning in LLM Agents

arXiv · Research · SCA

Restrospective: How Malicious Updates Poison Your Environment

ReversingLabs · Analysis · SCA

Agentic SDLC

From SELECT to SYSADMIN with SQL Copilot (CVE-2026-65669)

Embrace The Red · Analysis · Agentic SDLC

Aletheia: Permission-Minimality Testing for Coding-Agent Rules

arXiv · Research · Agentic SDLC

AI Coding Agents Exposed 13,000 Internal Images, Including Billing Records, on GitHub

The Hacker News · Research · Agentic SDLC

Remediation

PatchHolmes: Agentic Patch Retrieval via Listwise Selection

arXiv · Research · Remediation

Attackers Exploit Zimbra Flaw to Deploy Web Shells and Harvest Authentication Secrets

The Hacker News · Analysis · Remediation

Attackers Exploit NetScaler Flaw for Root Access, Deploy WHIPSHOT and SLAPSHOT

The Hacker News · Analysis · Remediation

Trust

Identity-Bound Governance Under Execution Uncertainty: An Accountability Proof Block for LLM Agent Persistent Halts, with Cryptographic Implementation and Cross-Model Calibration

arXiv · Research · Trust

Automatically Building Machine-Checked Assurance Cases from C Codebases to Requirements

arXiv · Research · Trust

Models

Do Defenses Against LLM Extraction Work Across Attacks? A Lifecycle Benchmark of Black-Box Model Extraction

arXiv · Research · Models

OpenAI reveals ‘novel’ encryption bypass used in distillation attack

CyberScoop · Analysis · Models

Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection

arXiv · Research · Models

AI Security

Sapien: A Stateful Policy Engine for Autonomous AI Agents

arXiv · Research · AI Security

Memetic Trojans: Social Contagions as Carriers of Adversarial Payloads in Agent Networks

arXiv · Research · AI Security

Measuring LLM Agent Defenses by Layer: GDA Measurement and an Envelope-Layer Attack-Defense Model

arXiv · Research · AI Security

Advisories in this window

  • 159 npm advisories · 143 critical · 9 high · 4 moderate · 3 low
  • 20 PyPI advisories · 4 critical · 7 high · 9 moderate
  • 6 crates.io advisories · 1 high · 3 moderate · 2 low
  • 2 Maven advisories · 2 high
  • 2 Packagist advisories · 1 high · 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.