CLAUDE CODE · DIRECT TRACK

Claude Code Mastery — The Tool, Plus What Powers It

Sixteen focused modules: from npm install through hooks, subagents, MCP servers, and the Agent SDK, plus the API foundations, prompt engineering, evaluation, RAG, and workflow patterns that make professional Claude Code work click.

For developers who want both depth and shipping speed: enough API knowledge to debug skill, subagent, and hook failures — without slogging through the full 30-module agent course.

Start with CC0 → or browse all 16 modules ↓
16modules
11–13 hrtotal
100%hands-on
Python + TSrunnable code

Why a separate Claude Code track?

The full 30-module course is a comprehensive agent-engineering curriculum. This track is purpose-built for Claude Code: every concept (API, prompts, RAG, evals, workflows) is anchored in skills, subagents, hooks, and MCP servers you'll actually ship.

What you get vs. what you skip

TopicDirect Track (this course, 16 modules)Full Agent Course (30 modules)
AudienceDevelopers using and extending Claude CodeEngineers building autonomous agents from scratch
Starting pointInstall → first session in 10 minMental models, tokens, LLM internals
Code examplesBash, JSON config, Python + TS for SDK usePython + TypeScript, Anthropic SDK, deep deployment
CoverageCLI features + API/prompts/eval/RAG/workflows tied to CC use casesAll of that plus M00 lifecycle, multi-agent systems, observability, capstones
End stateYou ship production CC workflows, write your own MCP servers, and build standalone agents with the Agent SDKYou build production agents end-to-end

Three commands to get rolling

# 1. Install (Node 18+)
$ npm install -g @anthropic-ai/claude-code

# 2. Authenticate (opens browser; use Pro / Team / Enterprise account or API key)
$ claude

# 3. Open it inside any repo
$ cd ~/your-project && claude

Detailed setup — including IDE extensions, login modes, and the /help tour — is the first half of CC0.

The 16-module curriculum

Linear path. Each module assumes the previous. Foundations → CLI features → tools & integrations → production patterns → building your own agents.

Setup
CC0 · 30 min Getting Started Install, authenticate, your first 5 minutes. The mental model: Claude Code is an orchestration framework, not a chatbot. beginnerinstall · /help · /init
Foundations — API & Prompting
CC1 · 45 min API Essentials for CLI Power Users Models (Opus/Sonnet/Haiku), Messages API, multi-turn, system prompts, temperature, streaming, structured output. Everything Claude Code hides under the hood. beginnerSDK · messages CC2 · 40 min Prompt Engineering for Claude Code Five techniques (clear, specific, XML tags, few-shot, thinking) applied to CLAUDE.md, skills, and subagent system prompts. beginnertechnique · structure
CLI Features
CC3 · 35 min CLAUDE.md & Memory The 5-tier configuration cascade, path-specific rules, AGENTS.md compatibility, and Auto Memory — how Claude remembers across sessions. beginnerconfig · auto-memory CC4 · 30 min Permissions & Sandbox Six permission modes, allow/ask/deny rules, and OS-level sandboxing. The safe defaults for dev · CI · production. intermediatesafety · modes CC5 · 45 min Skills & Slash Commands Build your own /deploy, /review, /triage. Frontmatter, $ARGUMENTS, shell injection, bundled skills, context: fork. intermediateskills · SKILL.md CC6 · 50 min Subagents Built-in (Explore, Plan, general-purpose), custom YAML, fork mode, persistent agent-memory, isolation worktrees, the @agent- mention. intermediate/agents · @agent- CC7 · 50 min Hooks All 28 lifecycle events, 5 handler types (command/http/mcp_tool/prompt/agent), async + defer permission, the /hooks menu. advancedpolicy · automation
Tools & Integration
CC8 · 55 min Tool Use Deep Dive The four-step tool-use loop, schemas, message blocks, multi-tool orchestration, fine-grained streaming, built-in text-edit and web-search tools. intermediatetool_use · schemas CC9 · 60 min MCP — Consumer & Server Add GitHub/Postgres/Slack/Jira to Claude Code, and write your own MCP server with tools, resources, and prompts. Inspector + protocol anatomy included. intermediateservers · FastMCP CC10 · 50 min Claude Features Extended thinking, prompt caching (90% cost cut), citations, image & PDF input, code execution + Files API. The features behind production cost & trust. intermediatecaching · citations
Quality, Knowledge & Patterns
CC11 · 45 min Evaluating Skills, Subagents & Prompts Test datasets, code-based and model-based grading, regression gates in CI. Make your prompts refactor-safe. intermediateeval · LLM-judge CC12 · 50 min RAG for Claude Code Chunking, embeddings, BM25, hybrid retrieval, exposing RAG as a subagent or MCP tool. Grounded answers over your docs. intermediateretrieval · hybrid CC13 · 50 min Workflows, Agents & Computer Use Chaining / parallelization / routing, the agent loop, environment inspection, and computer use (GUI automation). When to pick which. advancedpatterns · GUI
Production
CC14 · 60 min Power User & CI/CD Plan mode, two-Claude review, /loop, /schedule, worktrees, headless -p, GitHub Actions, image input, voice. The full operator toolkit. advancedworkflows · pipelines
Build
CC15 · 60 min Building Custom Agents with the Claude Agent SDK Same engine as Claude Code, exposed as a Python & TypeScript library. Two hands-on labs: a 5-minute "first agent" with built-in tools, then a UCC Filings Assistant with a custom MCP server wrapping the PublicRecords API. advancedSDK · Python · TypeScript

What you'll be able to do at the end

Configure Claude Code per repo

Hierarchical CLAUDE.md, path-specific rules, project-shared .claude/ files, gitignored local overrides, AGENTS.md compatibility.

🛡

Lock down what it can do

Permission modes, allow/deny rules, OS-level sandbox with filesystem & network walls, managed-policy settings for teams.

🔧

Build reusable workflows

Custom skills with shell-injected context, frontmatter-controlled invocation, subagents that learn across sessions.

Automate via hooks

Lint-on-write, refund-limit guardrails, central policy via HTTP hooks, deferred permissions, pipeline chaining.

🌐

Wire in — or build — MCP servers

GitHub, Postgres, Slack via official servers, plus your own FastMCP servers exposing tools, resources, prompts.

📖

Ground answers with RAG

Chunking, embeddings, BM25, hybrid retrieval — exposed as MCP tools so any Claude Code session benefits.

🔍

Evaluate before you ship

Test datasets, code & LLM-as-judge graders, regression gates in CI — make prompts refactor-safe.

🚀

Run it in CI/CD

Headless mode, two-Claude review, GitHub Actions, structured output via --output-format and --json-schema.

🤖

Build your own agents (SDK)

Same engine as Claude Code, exposed as a Python & TypeScript library. Define custom tools via in-process MCP, wire to existing APIs, ship standalone agents.

Need the full agent course?

If you're here to build autonomous agents end-to-end — deployment, observability, multi-agent systems, capstone projects — the full 30-module course is the right home. This Direct Track shares the M25/M26 reference material; the full course adds 28 more modules around it covering architecture, tracing, monitoring, and three domain-anchored capstones.