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 30 May 2026 · Retrospective · Covers 29 May 2026 00:00 to 30 May 2026 00:00 UTC

Read on the web

Top developments

ChatGPhish Vulnerability Turns ChatGPT Web Summaries Into a Phishing Surface

The Hacker News · Research · AI Security

Malicious Sicoob NuGet Steals Banking Credentials as npm Packages Target Cloud Secrets

The Hacker News · Research · SCA

CVE-2026-0257: Palo Alto Networks PAN-OS Authentication Bypass Vulnerability added to CISA's Known Exploited Vulnerabilities catalog

CISA · Advisory

SCA

Aligning on Machine-Readable Signals as the Foundation for Due Diligence

OpenSSF · Analysis · SCA, Trust

Typosquatted npm packages used to steal cloud and CI/CD secrets

Microsoft · Analysis · SCA

Agentic SDLC

What Breaks When LLMs Code? Characterizing Operational Safety Failures of Agentic Code Assistants

arXiv · Research · Agentic SDLC

Investigating Detection and Obfuscation of Prompt Injection Attacks Against Software Reverse Engineering AI Agents

arXiv · Research · Agentic SDLC, AI Security, Models

Trust

Bit-Exact AI Inference Verification Without Performance Tradeoffs

arXiv · Research · Trust

Models

BadBone: Backdoor Attacks Against Backbone Models in Visual Prompt Learning

arXiv · Research · Models

TRACE: Task-Aware Adaptive Self-Evolving Agentic Jailbreaking

arXiv · Research · Models

Persona Attack: Incremental Memory Injection Jailbreak Attack against Large Language Models

arXiv · Research · AI Security, Models

AI Security

From Prompt Injection to Persistent Control: Defending Agentic Harness Against Trojan Backdoors

arXiv · Research · AI Security

Depth-Dependent Indirect Prompt Injection in Tool-Calling ReAct Agents: Injection Depth, Payload Framing, and Turn-Budget Sensitivity

arXiv · Research · AI Security

Model Context Protocol SEP-2817: AI Invocation Audit Context in Request _meta

Model Context Protocol · Proposal · AI Security

Advisories in this window

  • 229 npm advisories · 218 critical · 6 high · 4 moderate · 1 low
  • 33 PyPI advisories · 11 critical · 14 high · 8 moderate
  • 17 Packagist advisories · 7 high · 10 moderate
  • 12 Go advisories · 5 high · 6 moderate · 1 low
  • 5 crates.io advisories · 1 high · 4 moderate
  • 2 NuGet advisories · 2 moderate
  • 2 Swift advisories · 2 moderate
  • 1 Maven advisory · 1 high
  • 1 Multi-ecosystem 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.