Anthropic published a security guide that basically tells you to stop trusting your own AI agents. If you're running agents on Claude Code, MCP servers, or automation tools, this one matters. Here's what it actually says: 👇

I think it is really worth reading this piece on RSI at Anthropic. There is a bit of navel-gazing, some marketing, and a lot of very sincere beliefs about what Anthropic thinks is likely in the near future of AI that you probably want to be aware of. anthropic.com/institute/recu…

Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor. It’s happening faster than we thought, and the implications deserve greater attention. anthropic.com/institute/recu…

Tokenmaxxing The stuff I'm hearing is just insane. People are spending hundreds of thousands of dollars a month on tokens? Guys, what are you shipping? I am seeing people fully enraptured by illusions of productivity. They have swarms of agents coordinated by Byzantine Octopus harnesses. They're munging thousands of tokens a second. They're doing all this stuff, churning unfinished marginalia faster than ever before. Spinning their wheels and shipping absolutely jack shit for their customers. Someone once pointed out that psychedelic drugs give their users the feeling of insight, but not actual insight. You wake up in the morning and you remember you had unlocked the mysteries of the universe, but all you can find are some spirals you scrawled on a notepad. Similarly, some engineers are falling all the way down the rabbit hole to the feeling of productivity, doing Heroic Doses of tokens on their way to True Productivity. Inevitably, they will reach psychosis before enlightenment. On a serious note, we're getting a lot of utility from AI for engineering at our company. I think we would really struggle to burn more than $5K per engineer per month. It turns out that extra-high-thinking mode on ultra-speed doesn't actually get you that much. Admittedly, we have a small codebase, but we have good tools -- and most of our problems are about thinking, understanding, and navigating real-world constraints. And I expect that's pretty much the same for most companies. Maybe if you're a big, established company with a 10M LOC codebase, you blow through the context much faster and can spend more money, but at that point -- do you really want to ship millions of lines of code a month? Aren't you concerned about stability? (I can think of a few valid exceptions, but for most F1000s this is just a non-starter.)

You outsourced your thinking and execution. Agents are killing that business model. medium.com/p/you-outsourced-…

Token economics are brutal for providers. Now enterprises are experiencing token-shock. medium.com/p/token-economics…

I just published Tokenmaxxing is useless, counterproductive… and, yes, stupid medium.com/p/tokenmaxxing-is…

The imminent token apocolypse (or when enterprises consume agents without subsides) medium.com/p/the-imminent-to…

Cloudflare explains why Mythos is so important (it is not necessarily what you think) medium.com/p/cloudflare-expl…

Token usage is skyrocketing, and so are costs medium.com/p/token-usage-is-…

Agent demand is accelerating, Token intensity is growing… something has to give… medium.com/p/agent-demand-is…

Your Agents are about to get much more expensive medium.com/p/your-agents-are…

Token Economics are Brutal (if you are a provider, that is) medium.com/p/token-economics…

Should we be doing real-time authorization for agents, with short-lived creds? Lengthy piece here makes the case ... nordicapis.com/why-real-time…

Every time I see a team celebrating their new "shared module," I remember this lesson. Reuse is a dangerous form of coupling. They found the same logic in two places and did what good engineers do: put it in one place and called it a win. Clean, responsible, textbook. Six months later, someone needs to change it. Suddenly, a small update for one team's requirements breaks three services, blocks two releases, and triggers an emergency meeting between people who've never talked to each other before. This is the cost nobody preaches about. DRY is one of those principles that feels unquestionably right until you apply it across team boundaries. The moment you share a module between domains, you're not just sharing code. You're creating a dependency that nobody owns and everyone resents. Before you reuse, ask: Will this change often? Does it belong to one domain? Are the consumers truly aligned in purpose? Will one team’s change surprise another team? If the answer to any of these is "I'm not sure," stop. Duplicate it. I know how that sounds. It feels lazy. It feels like the thing a junior developer does before they know better. But here's what nobody wants to say out loud: two independent implementations you control are almost always cheaper than one shared one serving masters with different goals. Duplication is a local problem. Coupling is an organizational problem. One of them you can fix in an afternoon. The other requires a meeting with five teams and someone's manager. Reuse isn't free. Treat it like the trade-off it is.

Each agent ships with the connectors, skills, and subagents the task needs, ready to use as-is or adapt to your firm's own standards. Read more: claude.com/solutions/financi…

On April 21, we challenged lawyers from all over the world to spend an afternoon building real, production-ready skills with Claude Cowork. 5 hours. 200 lawyers. 10 countries. 13 completed projects producing new legal AI skills. Every skill is live on our showcase site at go.heycounsel.com/hackathons

Eric Broda is a veteran of the software industry, and the co-author of the new O’Reilly book, Agentic Mesh: The GenAI-Powered Autonomous Agent Ecosystem. He joins @seanfalconer to discuss Agentic Mesh, distributed systems, and more. @ericmbroda softwareengineeringdaily.com…

If you are still thinking about how to build an agent, then you are missing the point. medium.com/p/if-you-are-stil…

The Task Boundary is the Enforcement Point for Agent Security and Governance medium.com/p/the-task-bounda…