Policy dependency / Stack layer
Retrieval that crosses into your dependencies' source, not just your repo, adds up to 6.3% pass@1 and survives version changes
arXiv 2609.09987
Policy dependency / Stack layer
CROSS-CATEGORY: Three Independent Agent-Action Gates Shipped in 48 Hours, All Judging the Command Against Stated Intent
Product Hunt, github.com/AGGIB/Stroq and rewarelabs.com (three independent sources; the 72% figure is Reware's own)
Stack layer / Contrast
Cognition ships SWE-2 on a Kimi K3 base: 92.8% on Terminal-Bench 2.1, and within a point of Fable 5.1 on FrontierCode at 64% lower cost
Cognition
Policy dependency / Contrast
A Fine-Tuned 4B Qwen in 2.6 GB Beats GPT-5.6 on a Transit-Kiosk Agent Benchmark, and PEFT Gains Vanish by 27B
arXiv 2609.10016
Threat pattern / Contrast
Deathray: a WebGPU shader freezes any M-series Mac from one link, and Apple won't patch it
auberon.xyz (224 points on HN)
Stack layer / Contrast
humans& Released Persimmon, a 550B User Simulator That Fools LLM Judges About 20% of the Time in Group Chats
humans&
Stack layer / Threat pattern
Google's Agent Development Kit for Python Carries a CVSS 10.0 Unauthenticated RCE via Test Session Replay
OffSeq Threat Radar
Stack layer / Update thread
SpecGuard Turns Speculative Decoding's Acceptance Rate Into a Free Runtime Backdoor Detector
arXiv