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

Weekly, week 6 of 2025 · Retrospective · Covers 28 January 2025 12:00 to 4 February 2025 12:00 UTC

Read on the web

Top developments

Google fixes Android kernel zero-day exploited in attacks

BleepingComputer · Incident · Remediation

How we estimate the risk from prompt injection attacks on AI systems

Google · Incident · AI Security

Constitutional Classifiers

Anthropic · Research · Models

8 Million Requests Later, We Made The SolarWinds Supply Chain Attack Look Amateur

watchTowr · Incident · SCA

Microsoft SharePoint Connector Flaw Could've Enabled Credential Theft Across Power Platform

The Hacker News · Research · Remediation

SAST

LLMSecConfig: An LLM-Based Approach for Fixing Software Container Misconfigurations

arXiv · Research · SAST

Fine-Grained 1-Day Vulnerability Detection in Binaries via Patch Code Localization

arXiv · Research · Remediation, SAST

AI Powered Remediation: Mend SAST Performs +46% Better Than Competitors

Mend.io · Analysis · Remediation, SAST

SCA

DeepSeek AI tools impersonated by infostealer malware on PyPI

BleepingComputer · Analysis · SCA

PyPI adds project archiving system to stop malicious updates

BleepingComputer · Analysis · SCA

Importing Phantoms: Measuring LLM Package Hallucination Vulnerabilities

arXiv · Research · SCA

OpenSSF Community Day NA 2025: Call for Proposals Now Open!

OpenSSF · Analysis · SCA, Trust

DAST

Towards Supporting Penetration Testing Education with Large Language Models: an Evaluation and Comparison

arXiv · Research · DAST

Agentic SDLC

Security and Quality in LLM-Generated Code: A Multi-Language, Multi-Model Analysis

arXiv · Research · Agentic SDLC

Remediation

Bill requiring federal contractors to have vulnerability disclosure policies gets House redo

CyberScoop · Analysis · Remediation

Broadcom Patches VMware Aria Flaws – Exploits May Lead to Credential Theft

The Hacker News · Advisory · Remediation

SoK: Towards Effective Automated Vulnerability Repair

arXiv · Research · Remediation

Vulnerability in popular AI developer could ‘shut down essentially everything you own’

CyberScoop · Analysis · Remediation

Trust

o3-mini vs DeepSeek-R1: Which One is Safer?

arXiv · Research · Trust

Models

PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling

arXiv · Research · Models

Adversarial Reasoning at Jailbreaking Time

arXiv · Research · Models

Detecting Backdoor Samples in Contrastive Language Image Pretraining

arXiv · Research · Models

A Framework for Double-Blind Federated Adaptation of Foundation Models

arXiv · Research · Models

AI Security

Firewalls to Secure Dynamic LLM Agentic Networks

arXiv · Research · AI Security

Contextual Agent Security: A Policy for Every Purpose

arXiv · Research · AI Security

Backline automatically remediates security vulnerabilities

TechCrunch: Security · Analysis · AI Security, Remediation

Advisories in this window

  • 222 npm advisories · 218 critical · 3 moderate · 1 low
  • 13 Maven advisories · 3 critical · 2 high · 7 moderate · 1 low
  • 12 Go advisories · 2 high · 10 moderate
  • 8 Packagist advisories · 2 high · 5 moderate · 1 low
  • 4 PyPI advisories · 1 critical · 1 high · 2 moderate
  • 4 crates.io advisories · 1 critical · 1 high · 2 moderate
  • 2 NuGet 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.