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 40 of 2025 · Retrospective · Covers 23 September 2025 11:00 to 30 September 2025 11:00 UTC

Read on the web

Top developments

Cisco warns of IOS zero-day vulnerability exploited in attacks

BleepingComputer · Incident · Remediation

Libraesva ESG issues emergency fix for bug exploited by state hackers

BleepingComputer · Incident · Remediation

SolarWinds releases third patch to fix Web Help Desk RCE bug

BleepingComputer · Advisory · Remediation

Malicious Rust packages on Crates.io steal crypto wallet keys

BleepingComputer · Analysis · SCA

Can We Trust AI To Write Vulnerability Checks? Here's What We Found

BleepingComputer · Research · Agentic SDLC

SAST

Red Teaming Program Repair Agents: When Correct Patches can Hide Vulnerabilities

arXiv · Research · Agentic SDLC, Remediation, SAST

CodeQL zero to hero part 5: Debugging queries

GitHub · Analysis · SAST

LISA Technical Report: An Agentic Framework for Smart Contract Auditing

arXiv · Research · SAST

BugMagnifier: TON Transaction Simulator for Revealing Smart Contract Vulnerabilities

arXiv · Research · SAST, SCA

SCA

Strengthening npm security: Important changes to authentication and token management

GitHub · Analysis · SCA

First Malicious MCP Server Found Stealing Emails in Rogue Postmark-MCP Package

The Hacker News · Research · AI Security, SCA

Binary Diff Summarization using Large Language Models

arXiv · Research · SCA

HFuzzer: Testing Large Language Models for Package Hallucinations via Phrase-based Fuzzing

arXiv · Research · SCA

DAST

Red Teaming Quantum-Resistant Cryptographic Standards: A Penetration Testing Framework Integrating AI and Quantum Security

arXiv · Research · DAST, Models

Intelligent Graybox Fuzzing via ATPG-Guided Seed Generation and Submodule Analysis

arXiv · Research · DAST

Semantic-Aware Fuzzing: An Empirical Framework for LLM-Guided, Reasoning-Driven Input Mutation

arXiv · Research · DAST

Agentic SDLC

Cross-Agent Privilege Escalation: When Agents Free Each Other

Embrace The Red · Analysis · Agentic SDLC, AI Security

Kicking off Cybersecurity Awareness Month 2025: Researcher spotlights and enhanced incentives

GitHub · Research · Agentic SDLC

Lita: Light Agent Uncovers the Agentic Coding Capabilities of LLMs

arXiv · Research · Agentic SDLC

Takedown: How It's Done in Modern Coding Agent Exploits

arXiv · Research · Agentic SDLC

Remediation

Urgent: China-Linked Hackers Exploit New VMware Zero-Day Since October 2024

The Hacker News · Analysis · Remediation

Salesforce Patches Critical ForcedLeak Bug Exposing CRM Data via AI Prompt Injection

The Hacker News · Advisory · AI Security, Remediation

Trust

Policy-as-Prompt: Turning AI Governance Rules into Guardrails for AI Agents

arXiv · Research · Trust

Operationalising AI Regulatory Sandboxes: Activities, Requirements, and Technical Assessment under the EU AI Act

arXiv · Research · Trust

Towards Quantum-Ready Blockchain Fraud Detection via Ensemble Graph Neural Networks

arXiv · Research · Trust

Update on the Next MCP Protocol Release

Model Context Protocol · Release · Trust

Models

Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on Graph Classification

arXiv · Research · Models

Fingerprinting LLMs via Prompt Injection

arXiv · Research · AI Security, Models

VAGUEGAN: Stealthy Poisoning and Backdoor Attacks on Image Generative Pipelines

arXiv · Research · Models

Agentic Specification Generator for Move Programs

arXiv · Research · Models

AI Security

Preventing Prompt Injection with Type-Directed Privilege Separation

arXiv · Research · AI Security

SecInfer: Preventing Prompt Injection via Inference-time Scaling

arXiv · Research · AI Security

When MCP Servers Attack: Taxonomy, Feasibility, and Mitigation

arXiv · Research · AI Security

SafeSearch: Automated Red-Teaming of LLM-Based Search Agents

arXiv · Research · AI Security, Models

Advisories in this window

  • 113 npm advisories · 87 critical · 7 high · 10 moderate · 9 low
  • 17 Maven advisories · 2 high · 14 moderate · 1 low
  • 12 Go advisories · 1 critical · 4 high · 4 moderate · 3 low
  • 9 PyPI advisories · 1 high · 6 moderate · 2 low
  • 4 NuGet advisories · 1 critical · 2 moderate · 1 low
  • 3 Packagist advisories · 1 high · 2 low
  • 2 GitHub Actions advisories · 1 critical · 1 high
  • 2 crates.io advisories · 2 moderate
  • 1 RubyGems advisory · 1 high
  • 1 Multi-ecosystem advisory · 1 critical

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.