
In production
Claude code for expert level debugging
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster through natural language commands.
Buildcamp learning paths
Learn how AI agents work in real applications. Explore agent development, tool connections, and practical guides to building with MCP.

Guide · 8 min read
Write one root AGENTS.md that Claude Code, Codex, and Cursor all read, then prove each agent summarises your shared project rules.

In production
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster through natural language commands.

Guide · 4 min read
Update Claude Code, sign in to the Claude Console and add an API key. Then ask Claude Code to build a Managed Agent from the Deep Researcher template, send it a…

Guide · 5 min read
Set up a beginner-friendly Supabase monitoring agent in Cursor. Learn MCP, read-only access, advisors, and a daily Generalist or Security check.

Guide · 4 min read
Connect Stripe’s official MCP (or drop test keys in .env.local), paste one prompt into Cursor or Claude, and take a test Checkout payment with card 4242.

Guide · 5 min read
Ask Grok Bot for a change in chat, let Cursor build it on your Supabase dev project through the Supabase MCP server, and add a safety rule that runs the advisors…

Guide · 5 min read
Agent Plugins are a portable way to give AI coding agents both knowledge and access

Guide · 4 min read
Pair the Cursor iOS app with your computer, keep it awake, start a small local agent task at your desk, then follow and reply from your phone. The agent stays on your…

Article · 5 min read
If you’ve heard of MCP servers and wondered what the hype is about—you’re in the right place.

Article · 11 min read
Three frontier models in 48 hours. A clear pick for Claude Code, Cursor, and Codex.

Article · 5 min read
A beginner-friendly guide to Anthropic's new agent infrastructure, and why it's the backbone of our latest course.