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

Daily, Tuesday 17 March 2026 · Retrospective · Covers 16 March 2026 00:00 to 17 March 2026 00:00 UTC

Read on the web

Top developments

GlassWorm Attack Uses Stolen GitHub Tokens to Force-Push Malware Into Python Repos

The Hacker News · Incident · SCA

CISA flags Wing FTP Server flaw as actively exploited in attacks

BleepingComputer · Incident

CVE-2025-47813: Wing FTP Server Information Disclosure Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory

SAST

PCodeTrans: Translate Decompiled Pseudocode to Compilable and Executable Equivalent

arXiv · Research · SAST

Securing our codebase with autonomous agents

Cursor · Analysis · Remediation, SAST

Why Codex Security Doesn’t Include a SAST Report

OpenAI News · Analysis · Agentic SDLC, SAST

SCA

Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework

arXiv · Research · SCA

DAST

Why Security Validation Is Becoming Agentic

The Hacker News · Analysis · DAST

Agentic SDLC

VibeContract: The Missing Quality Assurance Piece in Vibe Coding

arXiv · Research · Agentic SDLC

Remediation

UK’s Companies House confirms security flaw exposed business data

BleepingComputer · Analysis · Remediation

Trust

Governing frontier general-purpose AI in the public sector: adaptive risk management and policy capacity under uncertainty through 2030

arXiv · Research · Trust

New Microsoft Purview innovations for Fabric to safely accelerate your AI transformation

Microsoft · Analysis · Trust

Models

Evolving Contextual Safety in Multi-Modal Large Language Models via Inference-Time Self-Reflective Memory

arXiv · Research · Models

TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems

arXiv · Research · Models

SFCoT: Safer Chain-of-Thought via Active Safety Evaluation and Calibration

arXiv · Research · Models

AI Security

Amplification Effects in Test-Time Reinforcement Learning: Safety and Reasoning Vulnerabilities

arXiv · Research · AI Security

How Vulnerable Are AI Agents to Indirect Prompt Injections? Insights from a Large-Scale Public Competition

arXiv · Research · AI Security

Shadow AI is everywhere. Here’s how to find and secure it.

BleepingComputer · Guide · AI Security

Advisories in this window

  • 50 npm advisories · 39 critical · 5 high · 4 moderate · 2 low
  • 35 Go advisories · 2 critical · 10 high · 22 moderate · 1 low
  • 18 PyPI advisories · 2 critical · 12 high · 2 moderate · 2 low
  • 17 Packagist advisories · 1 critical · 7 high · 8 moderate · 1 low
  • 2 Hex advisories · 1 high · 1 moderate
  • 2 Maven advisories · 1 critical · 1 high
  • 1 crates.io advisory · 1 high
  • 1 Swift advisory · 1 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.