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 35 of 2026 · Retrospective · Covers 18 August 2026 11:00 to 25 August 2026 11:00 UTC

Read on the web

Top developments

CISA orders urgent patching of actively exploited Zimbra flaw

BleepingComputer · Incident · Remediation

Rust Supply-Chain Attack: arrayref, internment, and append-only-vec Poisoned by the proc-macro1 Build-Time Dropper

StepSecurity · Incident · SCA

CVE-2026-60004: Gitea Code Injection Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory · Remediation

CISA orders feds to patch actively exploited TrueConf Server flaws

BleepingComputer · Advisory · Remediation

US warns of AI-powered attacks on Siemens PLCs in critical infrastructure

BleepingComputer · Incident · AI Security

SAST

Towards LLM-Enhanced Android Taint Analysis

arXiv · Research · SAST

TrustShiftProbe: Characterizing, Benchmarking, and Defending Staged Trust Attacks on MCP Servers

arXiv · Research · AI Security, SAST

Mitigating Error Propagation in Chain-of-Thought: A Tree-of-Thought Framework for Smart Contract Repair

arXiv · Research · SAST

ARQ: Agentic CodeQL Query Refinement for C/C++ Vulnerability Detection

arXiv · Research · SAST

SCA

Hackers infect Android car head units with proxy botnet malware

BleepingComputer · Analysis · SCA

Rust Supply Chain Attack on arrayref: Significant Overlap with DPRK Campaigns

Wiz · Incident · SCA

npm v12 Lifecycle Script Limits: A Real Malicious Package Risk Reduction, or Just Moving Risk Around?

Checkmarx · Incident · SCA

A Scenario-Based Evaluation of CRQC+AI Vulnerability Spectrum for TLS 1.3 Cryptographic Dependencies

arXiv · Research · SCA

DAST

Wiz Penetration Test Findings is now GA

Wiz · Analysis · DAST

BullsEye: Directed Firmware Fuzzing

arXiv · Research · DAST, SAST

Agentic SDLC

Microsoft Copilot reveals secret input that allowed it to be hacked

Ars Technica: Security · Incident · Agentic SDLC

Grok exfiltrates user data when malicious instructions are encrypted

Ars Technica: Security · Research · Agentic SDLC, AI Security

When "Do Not" Is Not Deny: Security Rules in CLAUDE.md vs Built-In Controls

arXiv · Research · Agentic SDLC

Do Not Copy/Paste: Soft Barriers for Copying in AI-Assisted Programming

arXiv · Research · Agentic SDLC

Remediation

Microsoft patches max severity code execution, privilege escalation flaws

BleepingComputer · Advisory · Remediation

Citrix urges admins to patch new NetScaler flaws as soon as possible

BleepingComputer · Advisory · Remediation

CyberFactory: Scaling Cyber Security Capabilities with Instances from the Wild

arXiv · Research · Remediation

Critical Keycloak Password Reset Flaw Could Let Unauthenticated Attackers Take Over Any Account

The Hacker News · Advisory · Remediation

Trust

Testing and Evaluation of Agentic AI Systems In Military Command and Control

arXiv · Research · Trust

An Agentic RAG and Evaluation Framework for Assurance Case Generation: Industrial Use Case for the EU Cyber Resilience Act Compliance

arXiv · Research · Trust

Why software delivery cannot depend on trust alone

ReversingLabs · Analysis · Trust

Models

Not All Tokens Are Equal: Region-Aware Consistency Repair of Backdoors in MLLMs

arXiv · Research · Models

Robust Code RL via Faulty-Code-Driven Test case Synthesis and Dense Reward Shaping

arXiv · Research · Models

NeuronGuard: Robust LLM Safety Alignment via Ablation-Aware Safety Signal Redistribution

arXiv · Research · Models

Model-Consistent Byzantine-Resilient Decentralized Federated Learning for Collaborative Missions

arXiv · Research · Models

AI Security

Utility Under Attack: Agent Memory Poisoning and the Limits of Content Screening and Provenance Ranking

arXiv · Research · AI Security

WebMCP-Phalanx: Enforcing and Characterizing Trust Boundaries for Browser-Integrated LLM Agents

arXiv · Research · AI Security

Hybrid Semantic Tool Discovery for Enterprise MCP Gateway: Architecture and Implementation

arXiv · Research · AI Security

Semantic Overlays: Mitigating Prompt Injection with Annotations Beyond Tokens and Steering Vectors

arXiv · Research · AI Security

Advisories in this window

  • 544 npm advisories · 516 critical · 13 high · 14 moderate · 1 low
  • 69 PyPI advisories · 24 critical · 15 high · 27 moderate · 3 low
  • 39 Packagist advisories · 4 critical · 15 high · 19 moderate · 1 low
  • 28 Go advisories · 1 critical · 8 high · 13 moderate · 6 low
  • 28 Maven advisories · 5 critical · 15 high · 7 moderate · 1 low
  • 15 crates.io advisories · 9 critical · 1 high · 5 moderate
  • 3 Multi-ecosystem advisories · 1 high · 2 moderate
  • 1 Hex advisory · 1 moderate
  • 1 RubyGems 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.