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, Saturday 26 September 2026 · Retrospective · Covers 25 September 2026 00:00 to 26 September 2026 00:00 UTC

Read on the web

Top developments

MCP Security, Version 2.0: From Threat Taxonomy to a Model You Can Actually Audit Against

CoSAI · Research · AI Security

Case Study: How Does IBM Turn Open Source Participation Into Enterprise and Career Value?

OpenSSF · Research · Trust

Compromised GitHub Actions Came Back Online and Resumed Executing Mini Shai-Hulud Malware

The Hacker News · Incident · SCA

Cloudflare Fixes Flaw That Let One Container Read Another Customer's Leftover Disk Data

The Hacker News · Research · Remediation

SCA

Resource-Optimized and Energy-Aware Agentic AI Framework Anchored on Blockchain for Secure Software Supply Chains

arXiv · Research · AI Security, SCA, Trust

Agentic SDLC

Anthropic rolls out up to $250 in free Claude Code credits, but only for cloud sessions

BleepingComputer · Analysis · Agentic SDLC

MetaPermit: Scalable and Auditable Access Control for AI Agents via LLM-Inferred Meta-Attributes

arXiv · Research · Agentic SDLC, AI Security

Trust

Who Governs Data in the AI Era? A Computational Analysis of the U.S. Privacy Workforce in Job Postings

arXiv · Research · Trust

With the Rise of AI Agents, SOC 2 Should Adapt or Risk Irrelevance

BleepingComputer · Analysis · Trust

Models

TokenScanner: Detecting Backdoors and Discovering Triggers in Text-to-Image LoRAs via Full Vocabulary Scanning

arXiv · Research · Models

JevAdvBench: A Benchmark and Black-Box Attacks for Reinforcement Learning for Calibrated Decisions Models

arXiv · Research · Models

TempQ-Jail: Query-Constrained Candidate Ranking for Text-to-Video Jailbreak Attacks

arXiv · Research · Models

AI Security

Crypto-bound identity-verified capability tokens for coordinating distributed AI agents: A proposal

arXiv · Research · AI Security

Prompt Injection Detection for Email Agents Through Attack Chain Modeling

arXiv · Research · AI Security

On Anthropic’s AI Misuse Report

Schneier on Security · Analysis · AI Security

Advisories in this window

  • 18 npm advisories · 16 critical · 2 high
  • 6 PyPI advisories · 4 critical · 2 high
  • 4 Packagist advisories · 1 critical · 3 high
  • 3 Hex advisories · 2 high · 1 moderate
  • 2 NuGet advisories · 2 high
  • 1 Go 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.