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 37 of 2025 · Retrospective · Covers 2 September 2025 11:00 to 9 September 2025 11:00 UTC

Read on the web

Top developments

Google fixes actively exploited Android flaws in September update

BleepingComputer · Incident · Remediation

18 Popular Code Packages Hacked, Rigged to Steal Crypto

Krebs on Security · Incident · SCA

Salesloft Drift security incident started with undetected GitHub access

CyberScoop · Incident · SCA

Cloudflare hit by data breach in Salesloft Drift supply chain attack

BleepingComputer · Incident · SCA

Hackers steal 3,325 secrets in GhostAction GitHub supply chain attack

BleepingComputer · Incident · SCA

SAST

FuzzRDUCC: Fuzzing with Reconstructed Def-Use Chain Coverage

arXiv · Research · DAST, SAST

TraceLLM: Security Diagnosis Through Traces and Smart Contracts in Ethereum

arXiv · Research · SAST

SCA

s1ngularity's Aftermath: AI, TTPs, and Impact in the Nx Supply Chain Attack

Wiz · Incident · SCA

Model Namespace Reuse: An AI Supply-Chain Attack Exploiting Model Name Trust

Palo Alto Networks · Incident · SCA

Signal-Based Malware Classification Using 1D CNNs

arXiv · Research · SCA

AI-powered malware hit 2,180 GitHub accounts in “s1ngularity” attack

BleepingComputer · Analysis · SCA

DAST

FuzzBox: Blending Fuzzing into Emulation for Binary-Only Embedded Targets

arXiv · Research · DAST

Agentic SDLC

Red-Teaming Coding Agents from a Tool-Invocation Perspective: An Empirical Security Assessment

arXiv · Research · Agentic SDLC, AI Security

EchoLeak: The First Real-World Zero-Click Prompt Injection Exploit in a Production LLM System

arXiv · Research · Agentic SDLC, AI Security

Bypassing Claude Code: How Easy Is It to Trick an AI Security Reviewer?

Checkmarx · Analysis · Agentic SDLC

Remediation

CISA Orders Immediate Patch of Critical Sitecore Vulnerability Under Active Exploitation

The Hacker News · Analysis · Remediation

All You Need Is A Fuzzing Brain: An LLM-Powered System for Automated Vulnerability Detection and Patching

arXiv · Research · Remediation

Multi-View Slot Attention Using Paraphrased Texts for Face Anti-Spoofing

arXiv · Research · Remediation

"Abuse Risks are Often Inherent to Product Features": Exploring AI Vendors' Bug Bounty and Responsible Disclosure Policies

arXiv · Research · Models, Remediation

Trust

The Signalgate Case is Waiving a Red Flag to All Organizational and Behavioral Cybersecurity Leaders, Practitioners, and Researchers: Are We Receiving the Signal Amidst the Noise?

arXiv · Research · Trust

AI Governance in Higher Education: A course design exploring regulatory, ethical and practical considerations

arXiv · Research · Trust

BinaryShield: Cross-Service Threat Intelligence in LLM Services using Privacy-Preserving Fingerprints

arXiv · Research · AI Security, Trust

Constitutional Law and AI Governance: Constraints on Model Licensing and Research Classification

arXiv · Research · Trust

Models

Backdoor Attacks and Defenses in Computer Vision Domain: A Survey

arXiv · Research · Models

Are Targeted Data Poisoning Attacks as Effective as We Think?

arXiv · Research · Models

Mask-GCG: Are All Tokens in Adversarial Suffixes Necessary for Jailbreak Attacks?

arXiv · Research · Models

Multimodal Prompt Injection Attacks: Risks and Defenses for Modern LLMs

arXiv · Research · AI Security, Models

AI Security

Indirect Prompt Injection Attacks Against LLM Assistants

Schneier on Security · Incident · AI Security

SoK: Security and Privacy of AI Agents for Blockchain

arXiv · Research · AI Security

Parasites in the Toolchain: A Large-Scale Analysis of Attacks on the MCP Ecosystem

arXiv · Research · AI Security

Decoding Latent Attack Surfaces in LLMs: Prompt Injection via HTML in Web Summarization

arXiv · Research · AI Security

Advisories in this window

  • 36 npm advisories · 26 critical · 6 high · 3 moderate · 1 low
  • 21 Maven advisories · 2 critical · 3 high · 13 moderate · 3 low
  • 17 PyPI advisories · 4 critical · 5 high · 4 moderate · 4 low
  • 13 Packagist advisories · 2 high · 10 moderate · 1 low
  • 9 Go advisories · 2 critical · 3 high · 3 moderate · 1 low
  • 3 crates.io advisories · 1 high · 2 moderate
  • 2 GitHub Actions advisories · 1 high · 1 low
  • 1 NuGet advisory · 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.