Transition from "Point-and-Click" to "Command-and-Control."
Objective: Master the file system using the command line interface.
Estimated time: ~30 min. CLI fluency is non-negotiable for AI-assisted development. Your IDE has a terminal, your AI agent runs commands in it, and you need to read the output. This lab builds that muscle memory.
Prerequisite: Ensure you have your platform guide open:
Your first task is to build a professional project directory using only the terminal.
/AI-Research-Project
├── /data
├── /scripts
├── /outputs
└── project_notes.md
Commands to use: mkdir, cd, touch (or echo > for Windows).
Verify: Run ls (Mac) or dir (Win) inside AI-Research-Project. You should see 3 folders + 1 file. On Mac you can also try tree for a visual overview (install with brew install tree if needed).
We have "hidden" a research file in this repository (simulated).
.secret_ai_prompt?cat or type to read the contents of the main research draft.Commands to use: ls -la (Mac) or dir (Win), type/cat.
Tip: Hidden files start with a dot (.). On Mac/Linux, ls alone will not show them — you need the -a flag: ls -la.
Now, let's run a simple Python script that an AI agent wrote for us.
/scripts folder.vibe_check.py.print("Checking project status...")
print("AI Workspace: ACTIVATED")
python vibe_check.py.agents.md)This is the most important part of an agentic workflow.
agents.md in your project root.echo "# Current Task: Mastering CLI" > agents.mdecho "# Current Task: Mastering CLI" | Out-File -FilePath agents.mdA professional developer keeps a clean workbench.
/data folder (it was empty anyway).project_notes.md into /outputs.outputs to research_results.Commands to use: rmdir, mv or move.