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, Monday 3 August 2026 · Retrospective · Covers 2 August 2026 00:00 to 3 August 2026 00:00 UTC

Read on the web

Top developments

DenialRAG: Single-Document RAG Poisoning via Embedded Parametric Denial

arXiv · Research · AI Security

When Policies Change Probabilities: Modular Decision-Making for LLM Code Review

arXiv · Research · Remediation

No Single Neuron of Failure: Distributed Safety Alignment Against White-Box Attacks

arXiv · Research · Models

SCA

Vulnerabilities, Secrets and Misconfiguration in the Highest-Exposure Docker Hub Images

arXiv · Research · SCA

Agentic SDLC

Permission Denied: Policy-Graded Evaluation of Coding Agents in Hardened Environments

arXiv · Research · Agentic SDLC

Trust

Neuro-Symbolic Participation Governance for Verifiable AI Agents in Open Digital Twin Ecosystems

arXiv · Research · Trust

Models

Moving the Safety Barrier: Dynamic Routing Adaptive Alignment Against White-Box Attacks

arXiv · Research · Models

The Boy Who Cried Wolf: Adversarial Misclassification of Safe Inputs as Unsafe in Multimodal Guardrails

arXiv · Research · Models

FraudBench: Stress-Testing Policy-Grounded Banking Agents Against Adaptive Fraud

arXiv · Research · AI Security, Models

AI Security

Registry Descriptions Go Stale Unevenly: An 89-Day Measurement of Model Context Protocol Drift, and Why Drift-Ranked Re-Auditing Under-Covers It

arXiv · Research · AI Security

Advisories in this window

  • 8 npm advisories · 8 critical
  • 3 PyPI advisories · 2 critical · 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.