Resources
Playbooks, guides, and best practices for AI-native E2E testing.
What Is Claude Code? The Terminal Agent and Its Extension Model
Claude Code is Anthropic's terminal-based coding agent: it reads your codebase, edits files, runs commands, and connects to external tools over MCP. This hub explains how teams actually run it, how its five extension layers (skills, hooks, subagents, MCP servers, plugins) fit together, and the one thing it cannot see on its own: a real browser.
Claude Code Hooks: Deterministic Rules for an Agentic Workflow
Hooks are code that runs on Claude Code's lifecycle events: before a tool call, after an edit, when a task finishes. Here is the event model, the configuration format, and the recipes worth stealing, including test gates that hold every time.
Claude Code Plugins: The Bundle Layer Explained
Claude Code plugins package skills, subagents, hooks, and MCP server definitions into one versioned, installable unit. This guide covers plugin anatomy, marketplaces, the /plugin command, namespacing, and when a plain skill folder is the better choice.
Claude Code Subagents: How to Define Them and When to Delegate
Claude Code subagents are Markdown files with YAML frontmatter that run delegated tasks in their own context windows. This guide covers the file format, the built-in agent types, how delegation gets triggered, and three working recipes including a browser verification subagent.
Claude Code vs Antigravity: Terminal Agent or Agent-First IDE
Claude Code is a terminal agent extended through MCP, Skills, hooks, and subagents. Google Antigravity is an agent-first platform with an IDE, a CLI, and an Agent Manager for parallel work. This comparison walks the mechanisms: where each agent lives, which models it runs, how you extend it, and what each can verify.
How to Create a Claude Code Skill, Step by Step
A skill is one folder and one file away. This tutorial walks through creating a Claude Code skill: the SKILL.md format, the frontmatter fields that control loading, writing a description that actually triggers, and testing that it fires.