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

Read on the web

Top developments

Welcome to the Jungle: What We Found Inside 15,465 Public MCP Servers

The Hacker News · Research · AI Security

Ninja Forms plugin flaw exploited to hack WordPress sites

BleepingComputer · Incident

You Won’t Hear About These, Even In Myths (Atlassian Jira, Confluence (and more) Pre-Auth Arbitrary File Read CVE-2026-21589)

watchTowr · Analysis · Trust

SAST

Learning from Failures: A Failure-Driven Prompt Refinement for LLM-Based Vulnerability Analysis

arXiv · Research · SAST

CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?

arXiv · Research · Remediation, SAST

SCA

Large-scale Repository Engineering via Agent-Native Reusable Code Primitives

arXiv · Research · SCA

Agentic SDLC

SpecGuard: Proving a Task Is Broken Before the Agent Cheats

arXiv · Research · Agentic SDLC

A Case Study in Assuring AI-Written Software

arXiv · Research · Agentic SDLC

HarnessSecurity-Bench: Do Security Mechanisms Really Protect Coding Agent Harnesses?

arXiv · Research · Agentic SDLC

Remediation

CISO perspectives on managing vulnerability risks in the age of AI

Microsoft · Analysis · Remediation

Aikido Security achieves FedRAMP Moderate authorization

Aikido Security · Analysis · Remediation, Trust

Trust

OpenSSF Shares Expanded Membership and New Global Policy Resources During Community Day Europe

OpenSSF · Standard / policy · Trust

TwinGuard-Lite: A Rule-Based State-Admission Gateway for Generative Patient Digital Twins

arXiv · Research · Trust

Toward Alignment Scaling Laws: A Framework and First Preregistered Measurements

arXiv · Research · Trust

Models

Visual Memory Attacks Can Persist Through The KV Cache

arXiv · Research · Models

Secure Speculative Decoding for Large Language Models

arXiv · Research · AI Security, Models

ProximalFM: Amortized Proximal Causal Inference under Hidden Confounding

arXiv · Research · Models

AI Security

ASPIRE: Agentic Safety & Prompt Injection Red-teaming Engine

arXiv · Research · AI Security

AdvSim2Real : Training Web Agents Against Adaptive Prompt Injection in a Web World Model

arXiv · Research · AI Security

Surviving the Router: Optimizing Skill Injections for Retrieval and Execution

arXiv · Research · AI Security

Advisories in this window

  • 33 npm advisories · 15 critical · 10 high · 6 moderate · 2 low
  • 9 PyPI advisories · 2 critical · 1 high · 6 moderate
  • 4 Go advisories · 2 high · 2 moderate
  • 3 NuGet advisories · 1 critical · 1 high · 1 low

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.