💎 Key Takeaways
| Course Name | Claude Code 101 |
|---|---|
| Platform | Claude Academy |
| Price | Free |
| Duration | 1 hour, self-paced (12 lessons, 1 quiz) |
| Level | Beginner to Intermediate |
| Prerequisites | Basic familiarity with a code editor and command line; Claude Pro, Max, or Enterprise account, or API key |
| Skills | Agentic loop mechanics, Claude Code installation, prompt writing, context window management, CLAUDE.md configuration, MCP server integration, hook writing |
About
This course covers how to use Claude Code, Anthropic’s terminal-based AI coding agent, from setup through advanced customization. You learn the mental model first: what an agentic loop is, how the context window limits what Claude can see, and how tools and permissions shape what it can do. From there, the course builds into practical workflows including the Explore → Plan → Code → Commit rhythm for shipping real features, context management commands for long sessions, and code review with Claude Code.
The final section focuses on making the tool fit your specific setup. You write a CLAUDE.md file for persistent project memory, build custom subagents for repeated tasks, connect external data sources via MCP servers, and write hooks to enforce deterministic guardrails. The course targets both developers new to AI-assisted coding and experienced engineers who haven’t yet worked with an agent-based tool.
Who’s Teaching
N/A
What’s Covered
- What separates an AI coding agent from a chat-based assistant
- The agentic loop: context gathering, action, and result verification
- Installing Claude Code across terminal, VS Code, JetBrains, Claude Desktop, and the web
- Prompting in manual mode, auto-accept mode, and Plan Mode
- The Explore → Plan → Code → Commit workflow for feature development
- Context management using
/compact,/clear, and/context - Code review workflows using Claude Code
- Writing and maintaining a CLAUDE.md file for persistent project memory
- Building custom subagents and skills for repeated workflows
- Connecting external tools via MCP servers and writing hooks for deterministic control
Skills You’ll Develop
- Writing effective prompts for an agentic coding tool
- Managing context window state across long coding sessions
- Structuring AI-assisted development using a repeatable commit workflow
- Configuring Claude Code with project-specific memory and conventions
- Integrating external systems through MCP servers
- Building deterministic guardrails with hooks
Level
Beginner to Intermediate. No prior experience with AI coding tools is assumed, though basic comfort with a terminal and code editor is required. Experienced engineers new to agents will find the first-principles framing useful before moving into the customization sections.
