The threat model for AI agent skills is not the same as the threat model for traditional software packages. And the security tools catching up to that gap are finally starting to look like they belong in the same category. OpenClaw announced today that every ClawHub skill now ships with a Skill Card and is scanned by SkillSpector — an NVIDIA-developed tool built specifically to detect AI-native attack vectors inside agent skill files. ...

Something subtle but significant landed in OpenClaw v2026.6.2 today: agents can no longer write skills directly into production. Instead, they propose them — and humans decide whether those proposals become part of how the agent works from that point forward. That one change reframes a whole category of risk in autonomous agent systems. What Is Skill Workshop? Skills in OpenClaw are reusable procedural instructions agents load and execute. A skill might describe how to do invoice follow-up, how to run a repo health check, or how to process a specific kind of request. Once a skill is active, it shapes every future run that loads it. ...

Anthropic is scaling one of its most consequential and least-discussed programs. Project Glasswing — the company’s deployment of Claude Mythos Preview to help defend critical infrastructure — has just expanded to 150 additional organizations across 15+ countries, including India. This isn’t a product launch or a consumer announcement. It’s Anthropic quietly building a global AI-backed security layer around the systems civilization depends on. What Is Project Glasswing? Project Glasswing is Anthropic’s program to deploy Claude Mythos Preview — the most capable version of Claude in pre-release testing — specifically to organizations responsible for critical infrastructure defense. ...

The age of the solo AI coding assistant is ending. At Microsoft Build 2026, GitHub launched the GitHub Copilot App — a standalone desktop experience rebuilt from scratch for multi-agent coding workflows. And behind the scenes, a major model change is coming: Project Polaris, Microsoft’s in-house mixture-of-experts model, replaces GPT-4 Turbo as Copilot’s default starting August 2026. These aren’t incremental updates. They’re GitHub’s declaration that agentic software development is the default, not the exception. ...

Here’s a story that cuts through the Build 2026 announcements noise: Microsoft is canceling Claude Code licenses for thousands of its own engineers by June 30 and telling them to stop using the tool. The reason? Cost. Agentic coding tools are burning through enterprise AI budgets far faster than anyone projected. The decision comes from Microsoft’s Experiences + Devices division and represents a direct pivot: engineers who’ve been using Claude Code will now be directed toward GitHub Copilot CLI — Microsoft’s own product — as their primary AI coding assistant. ...

Microsoft just made its ambitions for always-on AI agents crystal clear. At Build 2026, the company unveiled Scout — a new category of AI agent it calls an “autopilot,” built natively on the OpenClaw agent framework and woven directly into Microsoft 365. This isn’t just another chatbot you ping and wait on. Scout runs in the background, constantly, understanding how work gets done across your apps and systems — and acting without needing to be prompted every time. ...

When people talk about “agent-first” software, they usually mean apps that have AI agents inside them. Microsoft is proposing something more radical: hardware designed from the chip up around the assumption that AI agents are the primary interface — not apps. That’s Project Solara, announced at Build 2026. And it’s built on Android, not Windows. What Is Project Solara? Project Solara is Microsoft’s new operating system platform for a new category of device — ones where AI agents replace apps as the way users get things done. Microsoft describes it as “a new platform built from the ground up to power agent-driven experiences.” ...

Anthropic Publishes Playbook for Running an AI-Native Engineering Org Most discussions of AI-assisted coding focus on the tool. Fiona Fung, Director of Engineering for Claude Code at Anthropic, is more interested in what happens to your organization when the tool works. Her talk at Code w/ Claude SF 2026 — now published as a written post on Anthropic’s blog — is one of the most honest accounts yet of what it actually takes to restructure an engineering team around agentic AI. Not what the marketing deck says. What breaks, what changes, and what you have to rebuild. ...

Cisco Cloud Control: One Platform for Humans and AI Agents to Run Enterprise IT The IT operations center of 2026 doesn’t look like the one from 2020. At Cisco Live US in Las Vegas this week, Cisco announced Cloud Control — a platform built explicitly for a world where human operators and AI agents run critical infrastructure together, not in parallel, not in handoff, but simultaneously. Cloud Control entered controlled US availability on June 2, the same day it was announced. This isn’t a preview — it’s a product launch. ...

Microsoft MDASH at Build 2026: When 100+ AI Security Agents Work Together The most interesting thing happening in AI security right now isn’t a better vulnerability scanner. It’s an orchestrated swarm of over 100 specialized AI agents that debate each other, challenge their own findings, and collectively hunt for bugs that any single model would miss. Microsoft’s MDASH — Multi-model Agentic Scanning Harness — just got significantly more powerful at Build 2026, and it’s a masterclass in what multi-agent architecture can accomplish when the problem is truly hard. ...