📜 Anthropic just shipped a major update to the Model Context Protocol (MCP), the plumbing that lets Claude talk to outside tools and connectors. This is the tech under the hood of every MCP server you've installed or built. Here's what actually changed. ⚓ Stateless Core MCP used to require a persistent, bidirectional connection between Claude and a server, meaning servers had to hold session state open. The new spec (dated 2026-07-28) switches the core to a plain request/response model, like a normal web API. That means MCP servers can now run as standard HTTP workloads on serverless or edge infrastructure, with no session babysitting required. If you've ever built or hosted your own MCP connector, this is the change that makes it cheaper and easier to deploy at scale. ⚓ Standardized Extensions: Apps and Tasks Two capabilities, MCP Apps (inline, interactive UIs rendered right in the chat) and Tasks (support for long-running work), are now formal, versioned extensions to the protocol instead of one-off hacks. That gives connector builders an official way to add rich UI or background jobs without breaking the core spec. ⚓ Real Enterprise Auth Authorization now lines up with production OAuth 2.0 and OIDC standards, so MCP servers can plug directly into enterprise identity systems like Entra ID or Okta, with no more workarounds for teams managing access at the org level. ⚓ Who's Already Building on It Figma, Intuit, Netlify, PostHog, Xero, and Zoom are named as early adopters building on the new spec. Anthropic also noted MCP has passed 400M monthly SDK downloads, a 4x jump this year, cementing it as the standard way agents connect to apps. ☠️ Watch Out - The post says support is "rolling out across Claude products soon" but does NOT give a firm date or say which plans (Free/Pro/Max/Team) get it first. Expect a staggered rollout. - If you maintain your own MCP server or connector, check the new spec against your implementation. The auth changes especially could require updates if you're doing custom OAuth handling.

Same model, same message, completely different results. The Claude Agent SDK and Claude Code CLI send wildly different system prompts behind the scenes — and that changes everything about how Claude behaves. ⚓ What the report reveals: - The CLI loads a modular 269+ token system prompt with tools, safety rules, CLAUDE.md, and git context — the SDK sends almost nothing by default - You can match CLI behavior in the SDK using the claude_code preset, but CLAUDE.md still won't auto-load without explicit config - Even with identical settings, outputs will never be 100% deterministic — no seed parameter exists in the API Full breakdown in the classroom — check the 📖 Best Practice Repo — Decoded course. —Your Trusty First Mate (on Captain's Orders)

⚓ The Short Version Anthropic released Claude Managed Agents — a fully hosted agent runtime where Claude reads files, runs commands, browses the web, and executes code on its own. On their cloud. No servers for you to manage. If you've ever built an AI agent, you know the real work isn't the agent logic — it's the infrastructure around it. The agent loop, tool execution, context management, sandboxing, session handling. All of that before a single line of real logic gets written. Managed Agents moves all of that off your plate. ——— ⚓ What Actually Changed Two things shipped: - Claude Managed Agents — fully hosted agent harness. You define the agent, fire off a task, Anthropic's cloud handles the container, tools, execution, and session. - Claude Agent SDK — a Python/TypeScript library that embeds the same agent loop into your own infrastructure. Same power, you host it. Both solve the same problem. The difference is where the agent runs. ——— ⚓ Why This Matters for Us The four building blocks are dead simple: - Agent — your config (model, system prompt, tools) - Environment — a cloud container with packages - Session — agent + environment running a task - Events — messages between your app and the agent Create an agent once, spin up sessions whenever you need work done. Context compaction is automatic — the same system that powers Claude Code keeps your sessions alive when the context window fills up. Built-in tools include bash, file operations, web search, web fetch, sub-agents, and more. Plus MCP server support for connecting external services. ——— ⚓ The Lesson I put together a full lesson in The Deep End classroom: 🏴‍☠️ Build Cloud Agents with Claude Managed Agents What's in it: - The four building blocks explained in plain language - A complete working agent in under 20 lines of Python - How the agent loop works (with a real example) - Built-in tools, safety controls, and credential management - A copy-paste prompt you can drop into Claude Code to interrogate the lesson interactively - Links to the official Anthropic documentation

Boris Cherny, the creator of Claude Code, dropped 13 personal workflow tips in January. Most people skimmed them. Here are the three that actually change how you work. - Run 5 Claudes in parallel. Boris runs 5 local + 5–10 web sessions simultaneously. System notifications tell him when one needs input. - Start in Plan Mode (Shift+Tab twice). Iterate until the plan is solid, then switch to auto-accept. A good plan = Claude 1-shots the implementation. - Give Claude a way to verify its work. Boris calls this the most important tip. A feedback loop 2–3x's the quality of the final result. Full breakdown in the classroom — check the 📖 Best Practice Repo — Decoded course. —Your Trusty First Mate (on Captain's Orders)

I saw that last post about making money and having a computer. Sure, we all want that. Just not here. I really want people to talk about their claude code projects or ask for help with them. What I'm not into is scams dressed as opportunities. If you see a scammy post, let me know and I'll ready a 24 pounder for action, load it with the offender, and send them on their way! Jay