Unlocking Gemini CLI with Skills, Hooks & Plan Mode

Greg Baugues — Fri, 20 Mar 2026 13:00:00 +0000

In Unlocking Gemini CLI with Skills, Hooks & Plan Mode, we moved past the basics and into the "power user" features of Gemini CLI.

I was joined by Jack Wotherspoon from the Gemini CLI team to show how developers can exert more control over their AI agents and handle complex, multi-step projects with confidence.
From a 20-minute app build to the introduction of a "read-only" research mode, this episode was packed with tools designed to bridge the gap between AI autonomy and developer intent.

The 20-minute build: From idea to deployment

To set the stage, Jack showcased Memory Wall, a digital bulletin board built using React, Three.js, and Firebase. The kicker? It took only 20 minutes to go from a blank slate to a live-deployed application.

This served as the playground for the day's deep dives:

1. Deterministic control with hooks

One of the biggest hurdles with AI agents is their non-deterministic nature. Hooks change that. They are scripts that run at specific lifecycle points—like at session start or before a tool call.

The "dev server" hook: Jack demonstrated a hook that checks if a local dev server is running on startup. If not, Gemini CLI alerts the user and offers to start it.
Safety first: You can use hooks to run linters or "security guards" that prevent the AI from writing messy code or deleting sensitive files.

Pro tip: Use the new Background Tasks feature (Control + B) to keep your dev servers running in the terminal without blocking your conversation with Gemini.

2. "Expert Hats": These specialized skills help to refine the behavior of these tools

If you’ve ever worried about "context bloat"—where an AI gets confused by too much information—skills are your solution. Jack described these as "library books on a shelf."

Progressive disclosure: Instead of loading every best practice into every prompt, Skills load specialized knowledge (like Three.js expertise or documentation style guides) only when they are triggered
The skill creator: Gemini CLI now has a built-in skill to help you build skills. Just ask: "Create a docs-writer skill for this project," and the CLI will walk you through a setup interview.

3. The "Ask User" tool

Gone are the days of the CLI just guessing what you want. With the new Ask User tool, Gemini CLI can pause and present interactive dialogues, multiple-choice questions, and yes/no prompts. This ensures the agent is aligned with your vision before it touches a single line of code.

4. Look before you leap with Plan Mode (preview)

Perhaps the most anticipated feature is Plan Mode, currently in preview. It transforms Gemini CLI into a read-only researcher.

Research first: In Plan Mode, the agent explores your codebase and external docs to create a structured "battle plan."
User approval: It presents this plan to you for feedback. Only once you give the "green light" does it switch to execution mode to start editing files.

Ready to dive deeper?

Watch: Missed the live demo? Catch the full replay here.

Learn: Take the free DeepLearning.ai course to get hands-on and learn more.
Contribute: Gemini CLI is open-source! Check out the "Help Wanted" labels on GitHub.

Getting Started with Gemini CLI

Greg Baugues — Wed, 11 Mar 2026 17:45:44 +0000

For Google Cloud's new weekly livestream series, we started with something every developer and knowledge worker is asking about right now: Generative AI, specifically for the command line.

While Gemini is great for chatting, bringing that experience into your terminal is a complete productivity game-changer. As host Stephanie Wong noted, the terminal isn't just for "hacker types," it's the fastest way to get things done without the layers of a browser or constant context switching.

In this episode, we were joined by Denise Kwan to break down the Gemini CLI stack and demonstrate how to turn your command line into a personal executive assistant.

Click here to install Gemini CLI

Here are the highlights, demos, and takeaways from Episode 1:

Stop alt-tabbing: Why the CLI?

The term "CLI" stands for Command Line Interface, and for developers, it represents the shortest path between an idea and an execution. Denise explained that by moving Gemini into the terminal, you remove the "heavy" layers of web applications. Instead of right-clicking through folders or uploading files one by one to a browser, you can interact with your entire local environment using natural language.

From zero to first prompt in 2 minutes

If you think setting up AI tooling is a weekend project, think again. Denise demonstrated that you can go from a fresh machine to a live prompt before your coffee gets cold.

By using npm and Node.js, the installation is a simple copy-paste job. Once installed, just type Gemini to launch.

Research at scale: Parallel search

This is the "a-ha!" moment for anyone doing market research or planning. Unlike the standard web interface that processes queries sequentially, Gemini CLI can run Google searches in parallel.

Whether you are comparing the latest electronics or doing competitive analysis for a product launch, the CLI fetches the "latest and greatest" from the live web simultaneously, allowing you to walk away and come back to a finished report.

The "Universal Plug": Understanding MCP

How does an AI agent talk to your local files, your databases, or your Google Drive? Enter MCP (Model Context Protocol). Denise described this as the "universal power adapter" for AI.
Instead of building custom integrations for every single tool, MCP provides a standardized way for Gemini to safely communicate with external systems. On top of that, Gemini CLI Extensions package these protocols with custom commands to make the agent even more reliable.

Demo: Vibe coding Tic-Tac-Toe app

We’ve all heard the buzzword, but Denise showed us what "vibe coding" actually looks like.

Taking a design spec from a Google Doc and a mockup image generated in AI Studio, she asked Gemini CLI to build a functional Tic-Tac-Toe web app from scratch. Because Gemini CLI is multimodal, she simply dragged and dropped the image into the terminal. The agent didn't just write the code—it identified its own errors during implementation and corrected them autonomously.

The "Morning Briefing"

Why should a non-developer care about Gemini CLI? Host Greg Baugues shared his secret weapon: the Morning Briefing.

By using the Google Workspace extension for Gemini CLI, he runs a custom command every morning that analyzes his pings, unread emails, and meeting notes from the previous day. It then cross-references his calendar to remind him of his highest-priority tasks. It’s a way to load your "personal context" so you can start your day focused on what matters.

Just get started!

To wrap up the episode, Denise challenged everyone to move past the "intimidation factor" of the terminal. The gap between a "what if" moment and a shipped product is now just a single prompt.

Don't just take our word for it: Gemini CLI has a free tier. You can install it, check the Terms of Service for your own peace of mind, and start vibe coding your own projects today.

Missed the episode? You can watch the full replay on-demand here.

Want to learn more? Join this short course to learn how to build real-world applications from the command line using Gemini CLI.

Forem: Greg Baugues