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 30 of 2025 · Retrospective · Covers 15 July 2025 11:00 to 22 July 2025 11:00 UTC

Read on the web

Top developments

CVE-2025-53770: Microsoft SharePoint Deserialization of Untrusted Data Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory · Remediation

Google fixes actively exploited sandbox escape zero day in Chrome

BleepingComputer · Incident · Remediation

CVE-2025-49704: Microsoft SharePoint Code Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory · Remediation

ExpressVPN bug leaked user IPs in Remote Desktop sessions

BleepingComputer · Incident · Remediation

Popular npm linter packages hijacked via phishing to drop malware

BleepingComputer · Analysis · SCA

SAST

FaultLine: Automated Proof-of-Vulnerability Generation Using LLM Agents

arXiv · Research · Remediation, SAST

LibLMFuzz: LLM-Augmented Fuzz Target Generation for Black-box Libraries

arXiv · Research · DAST, SAST

Architectural Backdoors in Deep Learning: A Survey of Vulnerabilities, Detection, and Defense

arXiv · Research · Models, SAST, Trust

SCA

NVIDIAScape - Critical NVIDIA AI Vulnerability: A Three-Line Container Escape in NVIDIA Container Toolkit (CVE-2025-23266)

Wiz · Analysis · SCA

GitHub abused to distribute payloads on behalf of malware-as-a-service

Ars Technica: Security · Research · SCA

Model Context Protocol SEP-985: Align OAuth 2.0 Protected Resource Metadata with RFC 9728

Model Context Protocol · Proposal · AI Security, SCA

Supply Chain Phishing Campaign Drops More Malware Into NPM: got-fetch 5.1

Checkmarx · Incident · SCA

DAST

Repeater Strike: manual testing, amplified

PortSwigger Research · Research · DAST

BACFuzz: Exposing the Silence on Broken Access Control Vulnerabilities in Web Applications

arXiv · Research · DAST

Harnessing LLMs for Document-Guided Fuzzing of OpenCV Library

arXiv · Research · DAST

LLAMA: Multi-Feedback Smart Contract Fuzzing Framework with LLM-Guided Seed Generation

arXiv · Research · DAST

Agentic SDLC

Multi-Stage Prompt Inference Attacks on Enterprise LLM Systems

arXiv · Research · Agentic SDLC, AI Security

Remediation

Max severity Cisco ISE bug allows pre-auth command execution, patch now

BleepingComputer · Analysis · Remediation

Microsoft: Windows Server KB5062557 causes cluster, VM issues

BleepingComputer · Analysis · Remediation

Microsoft mistakenly tags Windows Firewall error log bug as fixed

BleepingComputer · Analysis · Remediation

New Fortinet FortiWeb hacks likely linked to public RCE exploits

BleepingComputer · Analysis · Remediation

Trust

An Overview of the Risk-based Model of AI Governance

arXiv · Research · Trust

Small Changes, Big Trouble: Demystifying and Parsing License Variants for Incompatibility Detection in the PyPI Ecosystem

arXiv · Research · SCA, Trust

From Firms to Computation: AI Governance and the Evolution of Institutions

arXiv · Research · Trust

Coalition for Secure AI Marks First Anniversary with New Principles for Agentic Systems and Defender Frameworks

CoSAI · Analysis · Trust

Models

DeRAG: Black-box Adversarial Attacks on Multiple Retrieval-Augmented Generation Applications via Prompt Injection

arXiv · Research · AI Security, Models

AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning

arXiv · Research · Models

FedStrategist: A Meta-Learning Framework for Adaptive and Robust Aggregation in Federated Learning

arXiv · Research · Models

Fake or Real: The Impostor Hunt in Texts for Space Operations

arXiv · Research · Models

AI Security

SVAgent: AI Agent for Hardware Security Verification Assertion

arXiv · Research · AI Security

QSAF: A Novel Mitigation Framework for Cognitive Degradation in Agentic AI

arXiv · Research · AI Security

PromptArmor: Simple yet Effective Prompt Injection Defenses

arXiv · Research · AI Security

Manipulating LLM Web Agents with Indirect Prompt Injection Attack via HTML Accessibility Tree

arXiv · Research · AI Security

Advisories in this window

  • 179 npm advisories · 161 critical · 9 high · 7 moderate · 2 low
  • 15 Maven advisories · 2 high · 11 moderate · 2 low
  • 11 Go advisories · 6 high · 4 moderate · 1 low
  • 6 Packagist advisories · 3 critical · 2 high · 1 moderate
  • 5 PyPI advisories · 1 critical · 2 high · 1 moderate · 1 low
  • 2 GitHub Actions advisories · 2 high
  • 2 crates.io advisories · 1 high · 1 low
  • 2 RubyGems advisories · 2 moderate
  • 1 Multi-ecosystem advisory · 1 moderate
  • 1 NuGet advisory · 1 high

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.