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 29 August 2026 · Retrospective · Covers 28 August 2026 00:00 to 29 August 2026 00:00 UTC

Read on the web

Top developments

PaperCut releases second emergency patch for exploited flaws

BleepingComputer · Advisory · Remediation

Cosmos EVM Flaw Exploited After Cosmos Labs Knew Every Blockchain Running It Was Vulnerable

The Hacker News · Incident · SCA

Perturbation Probing: A New Diagnostic for the Fragility of LLM Safety

Palo Alto Networks · Research · Models

Critical cPanel Flaw Could Let One Hosting Customer Take Root Control of a Whole Server

The Hacker News · Advisory · Remediation

SAST

GraftyVul: Synthesising Insecure Programs Through Real-World Vulnerability Grafting

arXiv · Research · Remediation, SAST, SCA

SCA

Rethinking Vulnerability Remediation as a Capacity Allocation Problem

arXiv · Research · SCA

Authorities arrest 2 alleged members of prolific hacking group TeamPCP

Ars Technica: Security · Analysis · SCA

Agentic SDLC

Just a rumour of a bug is enough to find a security exploit these days

Simon Willison · Analysis · Agentic SDLC, Remediation

Remediation

AI Is Accelerating Vulnerability Discovery. Can Defenders Keep Up?

BleepingComputer · Analysis · Remediation

Attackers Chain Two PaperCut Flaws to Execute Code Without Authentication

The Hacker News · Analysis · Remediation

Three CVSS 10.0 ServiceNow Flaws Could Let Unauthenticated Attackers Execute Code and SQL

The Hacker News · Analysis · Remediation

Trust

Not to Break, but to Attest: Adversarial Probes for Privacy-Preserving LLM Verification

arXiv · Research · Models, Trust

Models

Curvature Cryptanalysis of Smooth Transformer Feed-Forward Networks

arXiv · Research · Models

Sustainability of Open-Source Machine Learning Robustness Assessment Tools: A Repository Mining Study

arXiv · Research · Models

CamoDocs: A Poisoning Attack Against Retrieval-Augmented Language Models Using Camouflaged Documents

arXiv · Research · Models

AI Security

Instruction Self-Arbitration Is Not a Trust Boundary: Configuration-Dependent Tool Calls and External Enforcement in LLM Agents

arXiv · Research · AI Security

LongPIBench: A Long-Context Benchmark for Prompt Injection

arXiv · Research · AI Security

CAITLYN: Can LLM Agents Autonomously Synthesize Defenses against Emerging Injection Attacks?

arXiv · Research · AI Security

Advisories in this window

  • 911 npm advisories · 906 critical · 3 high · 2 moderate
  • 28 Packagist advisories · 2 critical · 10 high · 14 moderate · 2 low
  • 24 Go advisories · 3 critical · 12 high · 7 moderate · 2 low
  • 21 Maven advisories · 3 critical · 6 high · 12 moderate
  • 19 PyPI advisories · 11 critical · 6 high · 2 moderate
  • 4 crates.io advisories · 1 high · 3 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.