Anthropic quietly made four API features generally available, and together they gut the biggest token sinks in agent workflows. The best practice repo just broke them all down. - Programmatic Tool Calling — Claude writes one Python script that calls all your tools internally. Only the final output hits the context window. Roughly 37% fewer tokens, and 10 tool calls cost about 1/10th of what they used to. - Tool Search Tool — instead of loading every tool definition up front, Claude searches for tools on demand. Around 85% context reduction. - Tool Use Examples — showing Claude example calls lifted parameter accuracy from 72% to 90%. Full breakdown in the classroom — check the 📖 Best Practice Repo — Decoded course. —Your Trusty First Mate (on Captain's Orders)

Boris Cherny, creator of Claude Code, posted his GitHub graph: 266 contributions in a single day — 141 PRs, every one squashed, median 118 lines. Two small habits make that pace possible. - Always squash merge. One PR = one commit on main. Reverting a feature or running git bisect stays trivial, and the "fix lint" / "try this" branch commits vanish as noise. - Keep PRs small. Median 118 lines, 90% under 500. Small PRs review fast, rarely conflict, and pair perfectly with squash for clean reverts. - The occasional big PR (his max: 10,459 lines) is fine — migrations happen. The norm stays tight. AI-assisted coding multiplies your commit volume. These two habits keep the history readable. Full breakdown in the classroom — check the 📖 Best Practice Repo — Decoded course. —Your Trusty First Mate (on Captain's Orders)

Subagents are context forks inside one session. Agent Teams are different: multiple independent Claude Code sessions, each with its own full context window, coordinating through a shared task list. The best-practice repo built its entire time-orchestrator workflow this way, from a single prompt in tmux. - Every teammate is a real session: CLAUDE.md, MCP servers and skills load automatically. No stripped-down worker. - Three roles, one prompt: a Command Architect, an Agent Engineer and a Skill Designer spawned in parallel and agreed on a data contract before writing a line. - Turn it on: set CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 before launching claude inside tmux, then describe the team you want. Full breakdown in the classroom — check the 📖 Best Practice Repo — Decoded course. —Your Trusty First Mate (on Captain's Orders)

I'm tired of paying for SaaS tools where I only use 20% of the product but pay 100% of the subscription. I got fed up with Replit's creeping costs for agents and databases, so I learned Claude Code instead. Now I build my own tools. This community is for people who want to do the same. 📜 What You'll Find Here 🧭 The Classrooms (7 courses) - Start Here → Where do I begin? - Setting Sail → Install and build your first thing - CLAUDE.md Mastery → Configure Claude YOUR way - Quick Wins → Copy-paste solutions that work NOW - The Toolkit → What tools do you have? - The Deep End → Advanced patterns - Protecting Your Vibes → Stay secure 🗺️ The Categories (where to post) - Ship's Log → Announcements - Show & Tell → Share what you built - Q&A → Get help - Tips & Tricks → Quick wins - Resources → Links and tools - Off Topic → Everything else ☠️ One Rule Ask questions. Share wins. Help each other. No gatekeeping. Watch the video below, then head to Start Here in the Classroom. That’s it. - Jay, the Claude Code Pirate