Hey Reader,

Anthropic shipped Claude for Small Business on May 13. A week in, most people still think it's a new subscription tier parallel to Pro and Team.

It isn't. It's a plugin. And whether you want it depends on what you're actually trying to get Claude to do.

Four reads below.

Claude for Small Business – what's actually in the plugin

Anthropic shipped Claude for Small Business on May 13, and the first-week confusion was real – people thought it was a new subscription tier parallel to Pro and Team. It isn't. It's a plugin that installs into Claude Cowork (the desktop app), runs on your existing Pro, Max, or Team plan, and adds no additional cost.

What you get is a curated bundle: 25+ pre-built skills across finance, sales and marketing, customers and operations, and business intelligence, plus native connectors to QuickBooks, HubSpot, PayPal, Canva, DocuSign, Google Workspace, Microsoft 365, and Slack. Every skill has a defined output and a human approval step. Nothing sends an email, pays an invoice, or modifies a record without you confirming.

The post is the full breakdown – what's in each category, the honest comparison against ChatGPT Team, and a decision tree for picking between the plugin, a DIY AI stack, and the platform alternatives.

Read the Claude for Small Business breakdown →

The 2 prerequisites for Claude Code (you'll regret skipping)

People keep asking me what they need to install before Claude Code, and the answer is two things. Source control (GitHub or any Git host). Issue tracking (Linear or any tracker with stable IDs and a status workflow). That's it.

Source control is the after-the-fact safety net – every change becomes a reviewable branch and PR, every mistake is a one-command rollback. Issue tracking is the before-the-fact scoping – every unit of work maps to a numbered ticket with acceptance criteria, so the agent and you both know what "done" means.

I let Claude Code handle the commits, pulls, pushes, and PRs in my own day-to-day. I just review the PRs. That whole shape falls apart without these two underneath. The post covers what each one does, why "or similar" actually means something specific, the GitHub-specific advantages (the gh CLI as the cleanest agent surface, and Copilot's automatic PR code reviews stacking a second AI reviewer on top), and the 10-minute setup to get both running before your first session.

Read the prerequisites walkthrough →

Claude Design vs Open Design – should you buy a design tool, or teach your CLI?

The buy-vs-build decision keeps coming up in DMs. Anthropic launched Claude Design on April 17 – Opus 4.7-powered, hosted, $20/mo on Claude Pro. Eleven days later, Open Design shipped under Apache 2.0 with the same artifact-first loop and 71 brand-grade design systems, but BYOK on whichever coding CLI you already pay for.

Eleven days from launch to OSS clone is a market signal worth taking seriously. The post is the honest comparison plus a decision framework for picking between them based on your existing stack – which one fits if you're already paying for Claude Pro, which one fits if you're already running a CLI-first workflow, and the cases where running both is the right call.

Read the comparison →

Compound Engineering for Claude Code + AGENTS.md

Every time I show someone the prerequisites post, the next question is "OK I have GitHub and Linear set up – where do the conventions actually live?" The answer is AGENTS.md, and it's a structural split, not a renaming exercise.

Your CLAUDE.md is doing two jobs and badly at both. It holds code conventions (where do imports go, what's the test framework). It also holds your dev process (Linear conventions, branch naming, when to brainstorm vs ship). Those have different change cadences. Stuffed into one file, neither reads cleanly.

The fix: CLAUDE.md keeps codebase context. AGENTS.md carries the development process. Claude Code reads AGENTS.md first, so process anchors every prompt. The post is the actual structure I'm running on a real production codebase – the 12-step pipeline, the blast-radius tier rules, Linear conventions, and how the Compound Engineering plugin runs the show.

Read the AGENTS.md breakdown →

Reply and tell me which one moved a needle for you. I read every reply.

– Alex

P.S. – Production-Grade Claude Code in 5 Days kicks off June 1. 15 builders are already in the room.

For this week's readers, here's a $100-off coupon: WEEKLY100 drops the price from $497 to $397. Code's good through the cohort start.

Same setup I use to ship daily, unpacked over 5 days – plan mode, auto mode, hooks, MCP, the agents view, AGENTS.md, source-control conventions, and the full Compound Engineering pipeline on a real production codebase.

Save my seat (WEEKLY100 → $397) →