Mastering the Claude Code CLI
$97one-time · lifetime
- 90-minute pre-recorded video
- CLI cheatsheet PDF
- .claude/ starter configs
- 5 reusable sub-agent definitions
- Hook recipes pack
- Lifetime replay + future updates
Pre-recorded class · 90 min
Mastering the Claude Code CLI.
Every flag. Every option. Every workflow. The CLI is the highest-leverage tool in your AI stack. Most people use 10% of it. This is the 90-minute deep dive that gets you to the other 90%.
- Course
- Intermediate
- 8 chapters
- On demand
$97 flat. No tiers, no upsell. Instant, lifetime access.
★★★★★“I am building things I never would have considered before making the investment in their products and classes.”John Wolf, on Trustpilot~/claude-cli, bashclaude-code
“the ideas can be applied immediately.”
“We now use 5 builds from DevLaunch”
“The support and follow-up has been above what I expected”
Sound familiar?
Most people use 10% of the CLI.
Prompts get longer instead of workflows getting smarter. Hooks, sub-agents, and MCP stay on the someday list. Every session starts with re-explaining the project.
If you have installed Claude Code, opened a project, and asked it a question, you are ready for this class. The goal is simple: stop copy-pasting prompts and start orchestrating.
The outline
The eight chapters.
One 90-minute video. Real screen recording with a terminal, real workflows, real flags annotated on screen.
Chapter 01
CLI mental model
- Modes, sessions, context, memory
- What actually persists vs. what's per-turn
- The 4 layers every Claude run touches
Chapter 02
Every flag worth knowing
- --print, --resume, --continue
- --output-format, --permission-mode
- --append-system-prompt and the 6 others nobody uses
Chapter 03
Slash commands & built-in skills
- /loop, /verify, /review, /init, /security-review
- When to invoke vs. when to inline
- Writing your own skill from scratch
Chapter 04
Hooks, in order
- SessionStart, UserPromptSubmit, PreToolUse, Stop
- Real recipes you can paste into settings.json
- The gotchas that brick a project for an hour
Chapter 05
Sub-agents
- Explore, Plan, general-purpose. When each wins
- Writing custom sub-agents (the YAML frontmatter)
- Parallel invocation patterns
Chapter 06
.claude/settings + CLAUDE.md
- What belongs in project memory vs. user memory
- Permissions allowlist patterns
- The auto-memory system, properly configured
Chapter 07
Headless Claude
- Piping with bash, JSON streaming
- CI integration patterns
- Scheduled agents and the cron primitive
Chapter 08
Demo. Real workflow end to end
- Watch a real task from prompt to PR
- Every flag and hook used annotated on screen
- The workflow we run daily, recorded once, yours forever
A straight answer
Who this is for, and who it is not.
This fits if
- You use Claude Code daily but still drive it like a chat box
- Your prompts are getting longer instead of your workflows getting smarter
- You've heard of hooks / sub-agents / MCP but never wired them up
- You're tired of re-explaining your project to Claude every session
- You want one source of truth for what every flag actually does
It is not the right fit if
- You've never opened a terminal
- You're looking for a no-code AI tool
- You want a course on prompt engineering theory
Your instructor
Taught by Aaron Rountree
Aaron builds software and runs his own business with the tools he teaches. DevLaunch is where he records the work: client portals, content systems, AI employees, and the workflows that hold them together.
You see how the pieces fit, so you can use the same approach on your own projects.
What ships
Everything ships at $97.
No pro tier. No premium upsell. One price, everything in.
./includes/
├── video/
│ └── mastering-claude-cli.mp4 90 min · lifetime replay
├── cheatsheet/
│ └── claude-cli-cheatsheet.pdf every flag · printable
├── .claude/
│ ├── settings.json starter project config
│ ├── settings.local.json user-level overrides
│ └── CLAUDE.md project-memory template
├── sub-agents/
│ ├── explore.md fast read-only search
│ ├── plan.md architect / planner
│ ├── reviewer.md independent code review
│ ├── verify.md run-the-app verification
│ └── researcher.md web + doc synthesis
└── hooks/
├── SessionStart.example.json
├── UserPromptSubmit.example.json
├── PreToolUse.example.json
└── Stop.example.json
5 directories · 14 files · 1 lifetime licenseCustomer experiences / Trustpilot
What customers say
about DevLaunch.
1 / 5
Selected excerpts from reviews of DevLaunch training, software, and services. Individual customer experiences across DevLaunch.
Single tier · everything included
One price. Everything in.
Secure checkout via Whop. Stripe and PayPal accepted. Instant access.
Questions
Before you join.
Is this live or recorded?
Pre-recorded and on demand. Watch tonight, watch Saturday morning, watch six months from now. Lifetime replay.
What's the format?
One 90-minute video, structured as eight chapters. Real screen recording with terminal, real workflows, real flags annotated on screen. Includes a downloadable cheatsheet PDF and the .claude/ starter pack.
Do I need to be a developer?
You need to be running the Claude Code CLI already. If you've installed it, opened a project, and asked Claude a question, you're ready. This isn't an intro to programming.
Will this work for me if I'm on Pro / Max?
Yes. Everything covered works on every Claude Code subscription tier. The class covers Agent SDK credits and which workflows use which.
What if I already know this stuff?
If you can already explain the difference between SessionStart and UserPromptSubmit hooks, and you've shipped a custom sub-agent, this might not be for you. Otherwise, there is a good chance a lot of this is new to you.
Ninety minutes. Ninety-seven dollars.
Stop driving Claude like a chat box.
Every flag, every hook, every workflow. Recorded once, yours forever.
Get the class · $97 Instant access
