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, Friday 9 October 2026 · Retrospective · Covers 8 October 2026 00:00 to 9 October 2026 00:00 UTC

Read on the web

Top developments

Tensorlake npm Package Compromised: A Worm With a Hostage Token That Wipes Your Machine If You Revoke It

StepSecurity · Incident · SCA, Trust

How one bug bounty researcher chooses the features they investigate

GitHub · Research · Agentic SDLC, Remediation

ARTEX AI Pentesting Tool Used in Data Theft Attacks on South Korean Financial Firms

The Hacker News · Incident · DAST

SAST

ObliVul: Alert-Conditioned Safety Obligation Modeling and Bidirectional Counterfactual Validation for Code Vulnerability Detection

arXiv · Research · SAST

SCA

EIFL: Efficiently Protecting Global Model Privacy and Integrity Against an Untrusted Server in Federated Learning

arXiv · Research · SCA

Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub

arXiv · Research · Agentic SDLC, SCA

Agentic SDLC

SWE-Duel: Towards a Self-Scaling Arena for Coding Agents

arXiv · Research · Agentic SDLC

Where Do the Tokens Go? Understanding and Reducing Costs in LLM Agents for Vulnerability Discovery

arXiv · Research · Agentic SDLC

Trust

Top FedRAMP-compliant security tools in 2026

Aikido Security · Analysis · Trust

Models

Cancerous Backdoor Attack: Gradual Activation via Invisible Tokens and A Stage Controller

arXiv · Research · Models

One Word Opens the Gate: The Option-Channel Attack on Typed Decision Models as Agent Guardrails

arXiv · Research · Models

AI Security

StepSecurity Now Inventories AI Agent Skills in Your GitHub Repositories and on Developer Machines

StepSecurity · Analysis · AI Security

From Reactive Containment to Proactive Assurance: Lessons from OpenAI, Anthropic, and Google Agent Security Incidents

arXiv · Research · AI Security

LTBD: Learnable Trust-Boundary Delimiters for Prompt Injection Defense

arXiv · Research · AI Security

Advisories in this window

  • 99 npm advisories · 73 critical · 11 high · 14 moderate · 1 low
  • 40 PyPI advisories · 4 critical · 15 high · 19 moderate · 2 low
  • 17 Go advisories · 7 high · 10 moderate
  • 10 Maven advisories · 2 critical · 4 high · 3 moderate · 1 low
  • 3 Packagist advisories · 3 moderate
  • 2 RubyGems advisories · 2 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.