Install Gemini CLI

Introduction

Gemini CLI has been one of the most popular AI agents in the first half of 2025. It’s similar to Claude Code, bringing its power directly into your terminal. Although other terminal AI agents exist, their pricing plans are quite different. Gemini CLI provides a free tier with 100 requests per day using Gemini 2.5 Pro, and you can unlock Tier 1 by upgrading to a paid plan.

Prerequisites

I’m going to use npm to install Gemini. My Node.js version is v24.4.1, and my npm version is 11.4.2.

Gemini needs Node.js version 20 or higher installed.

If you’re using macOS, you can also choose Homebrew to install the Gemini CLI.

Installation

Now, let’s install it using npm. After installation, you can run gemini directly in your terminal.

npm install -g @google/gemini-cli

I’m using the Use Gemini API key authentication method, so I need to generate a key from Google AI Studio and set it in .zshrc (or .bashrc) by adding this line:

export GEMINI_API_KEY=YOUR_API_KEY

And then you can try Gemini now!

Run some examples

Prompt: give me suggestions for the socket functionality of this project?

Response:

✦ Thank you for sharing that. The architecture document is very clear. Using Clean Architecture with FastAPI is a solid foundation.

Given that this is a control plane for a load balancer, the most valuable use for WebSockets is to provide real-time observability and interactivity. Here are some specific suggestions for socket
functionality that would fit well with your existing architecture:

1. Real-time Health Check Monitoring
   Your system manages services and endpoints. A crucial next step is to know if they are online and healthy.

- How it would work: You could create a WebSocket endpoint like /ws/status. A background worker in your backend could periodically perform health checks on all registered ServiceConfigs. When a
  service's status changes (e.g., from "healthy" to "unhealthy"), the server pushes a JSON message to all connected clients (e.g., a UI dashboard).
- Benefit: Administrators get instant visual feedback on the state of their services without needing to constantly refresh a page or manually check endpoints.

2. Live Configuration Event Broadcasting
   When multiple users are managing the system, it's important that they all see the most current state.

- How it would work: After any use case that modifies a ServiceConfig, Group, or Endpoint (Create, Update, Delete), you could broadcast an event message over a general-purpose WebSocket like
  /ws/events. The message could specify the entity type and the change that occurred.
- Benefit: Any connected UI can listen for these events and intelligently update its state, ensuring all concurrent administrators have a consistent view of the system's configuration.

3. Real-time Log Streaming
   For debugging purposes, streaming logs from services in real-time is incredibly powerful.

- How it would work: You could implement a WebSocket endpoint like /ws/logs/{service_id}. When a client connects, the backend would start streaming logs for that specific service (e.g., by tailing a
  log file or using the Docker logs API for ContainerService).
- Benefit: Provides a powerful, integrated debugging tool directly within your management application, saving developers time and effort.

4. Live Traffic & Performance Metrics
   Since this is a load balancer API, providing insight into the traffic it's managing is a natural fit.

- How it would work: This is more advanced, as it would require your actual load-balancing data plane to push metrics (e.g., requests per second, error rates, active connections) to your API backend,
  perhaps via a message queue like Redis. Your FastAPI app could then broadcast these metrics over a WebSocket like /ws/metrics to be displayed on live charts.
- Benefit: Delivers a real-time dashboard for monitoring application performance and health, which is a key feature for any serious load balancer.

FastAPI has excellent native support for WebSockets (https://fastapi.tiangolo.com/advanced/websockets/), making the implementation of these features quite straightforward within your existing
structure. You could add a new websockets directory in the presentation layer for each module.

Would you like me to elaborate on any of these suggestions or perhaps draft a small code example for one of them, like the health check monitor?

Conclusion:

The Gemini installation is very simple. Although I am using Neovim with Avante, Gemini gives me more power to use the terminal. Next, I will explore how to use Gemini with an MCP server and integrate the workflow into my daily tasks.