News from the agent stack.
GPT-5.6 Just Made Itself Cheaper
OpenAI cut GPT-5.6 Luna's price by 80% — and the gains came from pointing GPT-5.6 Soul at its own infrastructure. Matthew Berman on the 20% serving-cost drop the model found for itself, the Karpathy auto-research parallel, why the flagship conspicuously didn't get cheaper, and what recursive self-improvement means for everyone trying to catch up.
The Next Bottleneck Is Understanding the Business
A forward-deployed engineering talk arguing that once execution stops being the constraint, specification becomes the scarce input — the undocumented exception handling where process time actually goes. A customer who spent $5 million and five years migrating to NetSuite defines the architecture. The most credible moment is a vendor admitting frontier models produce verbose, incorrect analysis because they lack editorial judgement — which contradicts the CEO's claim three minutes earlier that knowledge work is almost solved.
Jensen Huang: Why Companies Need Open Agent Systems
NVIDIA CEO Jensen Huang and Harrison Chase (LangChain) discuss why companies need domain-specific super agents on open harnesses. They announce Deep Agents + Nemotron 3 Ultra inside OpenShell — a secure open runtime. Nemotron 3 Ultra hits 86% vs Opus 87% at 10x lower cost.
AI Has Hacked the Code of Human Civilization — Yuval Noah Harari
Historian Yuval Noah Harari argues in his 2026 Oxford Tanner Lecture that AI is not a tool but an agent, and its natural habitat is the bureaucratic systems humans built using language — making it uniquely positioned to take over law, finance, religion, and even personal relationships from within.
Linus Torvalds Speaks on AI's Flood of Bugs & Linux's Future
In a fireside chat at the Open Source Summit, Linus Torvalds reveals that AI tools have driven a 20% spike in Linux kernel commits, flooded the security mailing list with duplicate reports, and killed security through obscurity — but he remains optimistic that finding bugs beats not finding them.
Most Enterprise Agentic Projects Are Doomed, Here's Why — Jess Grogan-Avignon & Jack Wang, Accenture
Jess Grogan-Avignon and Jack Wang from Accenture deliver a brutally honest talk on why most enterprise agentic projects fail — and it has nothing to do with the technology. They built the app in 2 weeks; getting it to production took 6 months.
A Formal Language for Writing Agent Instructions: Matt Pocock's Skills v1.2.0
A twelve-minute walkthrough of five new skills and a major infrastructure update for the most-starred agent skills repository on GitHub — now 204K stars, MIT-licensed, and an official Claude Code marketplace plugin alongside new Codex sidecar files. The headliner is /wait-what for Opus verbosity, /grill-me with multi-question dependency graphs, and /wizard for deterministic provisioning — but the skill nobody called exciting is the one worth reading: /writing-for-agents, an 81-line formal language for designing instructions agents consume — context pointers, the two loads, information hierarchy, completion criteria, leading words, negation as a failure mode, and pruning.
Forgetting as the Compression Step: Tencent's Agent Memory Engine, Checked
Tencent open-sourced an MIT agent-memory plugin that improves results by throwing information away: verbose tool logs fold into a Mermaid graph with node_ids you can grep back, and long-term memory climbs an L0-to-L3 pyramid modelled on Tulving's 1972 consolidation arrow. The repo is real (10,811 stars), the architecture is documented, and the benchmark table publishes raw token counts - so the presenter divides them on camera and finds one row that does not reconcile: SWE-bench prints -33.09% but computes to -31.63%. He also flags that every number is vendor-produced with no independent reproduction, and names the open GitHub issue where prompt-cache invalidation claws back part of the saving. The gains shrink as tasks get harder: +51.52% on WideSearch, +9.93% on SWE-bench. The video's own description gets this wrong; the video does not.
Wayfinder: Planning as a Map You Can Leave and Return To
Most agent planning tools die at the edge of a context window, and the real cost is not lost notes but trimmed ambition. Matt Pocock's Wayfinder skill moves the plan out of the session and into the issue tracker as a map of decision tickets, so work outlives any single conversation. Reading the skill source alongside the video surfaces two constraints the video never states: every ticket is either human-in-the-loop or agent-driven, and the skill forbids an agent standing in for the human's side of a discussion — the rule that keeps a decision map from looking complete while encoding nobody's judgement.
This Skill Just Made Kimi K3 A 10x Better Designer
AI LABS tests Kimi K3's frontend design capabilities against Claude and GPT, then introduces Hallmark — an anti-AI-slop design skill with 58 detection gates, 100+ references, and four verbs (default, audit, redesign, study) that strips every model's default patterns and produces intentional, original designs.
System Prompts Leaks GitHub Explained: What AI Agent Prompts Reveal
Alex Hitt breaks down the system_prompts_leaks repository — 43,600+ stars on GitHub — revealing how researchers verify leaked AI system prompts, the four-layer architecture inside commercial models, and why prompt obfuscation as a competitive moat is dead.
Don't Waste Time on Specs — /prototype Instead
Matt Pocock argues that spec-driven AI development wastes effort. His /prototype skill creates throwaway code artifacts at higher fidelity than discussion, letting you iterate on live UIs and logic before handing off to production — bridging the gap between spec and working software.
A Global Workspace in Language Models
Anthropic discovers that Claude developed an internal J-space — a privileged mental workspace analogous to the global workspace theory of conscious access in neuroscience. It emerged on its own during training, enables silent reasoning, and can be used to detect hidden goals and deception.
The New SDLC With Vibe Coding
A comprehensive whitepaper by Addy Osmani, Shubham Saboo, and Sokratis Kartakis exploring how AI transforms the software development life cycle — from casual vibe coding to disciplined agentic engineering, context engineering, the factory model, and the economics of AI-driven development.
How Your Agents Can Write and Optimize Their Own Skills
A deep dive into automated skill optimization frameworks — SkillOpt, GEPA, and EvoSkill — that treat agent skill files like trainable neural network parameters, enabling agents to iteratively debug, validate, and update their own skills without touching model weights.
Copilot Cowork Is Now Generally Available — Anthropic Claude Meets Microsoft 365
Microsoft's Copilot Cowork — the deep-work AI assistant powered by Anthropic Claude — is now GA for all Microsoft 365 Copilot users. Brings model choice (Opus 4.8, Sonnet 4.6, GPT 5.5), partner plugins (Monday.com, Miro, Adobe, Atlassian), and Edge browser access.
LangChain Deep Agents — The Batteries-Included Agent Harness, Explained
LangChain open-sourced Deep Agents — an opinionated, batteries-included agent harness built on LangGraph. Inspired by Claude Code, it ships with planning, filesystem access, sub-agent orchestration, context management, and persistent memory out of the box.