Claude Code Tips and Tricks
Everything you need to get started with Claude Code. Best practices, tips, and tricks from my experience using it.
The 20% that gets you 80% of the way there. Distilled, actionable guides that actually move the needle for your business.
Everything you need to get started with Claude Code. Best practices, tips, and tricks from my experience using it.
A practical mental model for Claude's two most-confused building blocks, with a worked agency example and tips for keeping your work portable.
A design philosophy for Claude Code skills: script every step a unit test could verify, and reserve the LLM for the judgment calls only it can make.
How to turn VS Code into a proper markdown reading and writing environment with mode switching, fonts, and Chrome preview
A simple filesystem trick that lets you and your AI agent co-maintain product documentation in real time
Get an AI coworker in your Slack that learns about you, searches the web, and works with any model you choose.
What the orchestrator skill does under the covers on Claude Code: named sessions that message each other, worktrees built on the right branch, and the permission settings that can hold a message.
What the orchestrator skill does under the covers on Codex: everything routes through one main thread, subagents can't nest, and a few settings decide how many workers you can keep open.
How I run a build that's too big for one agent session: one orchestrator plans, sequences, and merges, one worker per slice builds in its own worktree, and I make the go/no-go calls.
How to organize a knowledge vault for consulting work using PARA methodology, client subfolders, and Obsidian's new CLI for safe moves and renames
This guide has moved. The Playwright MCP vs CLI comparison now lives in Part 3 of the browser-automation series.
A structured workflow for producing clean code with AI agents, featuring skills for feature impact analysis and automated documentation
Get started with text expanders you can use anywhere on your system