The terminal is already a powerful tool, but the right command-line utilities can make it even better. Here are a few essential tools I use daily that you might find helpful too.

HomeBrew

Homebrew is a package manager for macOS (and Linux) that lets you easily install, update, and manage software and developer tools from the command line.

# Installation command
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

After running the command, restart your terminal, and you’re good to go. Before we start installing, let’s understand the difference between a “formula” and a “cask” in Homebrew.

Formula: Command-line tools, libraries, and languages (e.g., Kotlin, Go). No special flags are needed when using brew commands with formulas. All Formulas

Cask: GUI applications like Chrome, VSCode, etc. These require a --cask flag for brew operations. Cask Directory

brew install git             # Installing a formula
brew install --cask firefox  # Installing a cask

brew uninstall foo           # Remove a package
brew upgrade foo             # Upgrade a specific package
brew list                    # See all installed packages
brew search foo              # Find available packages
brew info foo                # Get details about a package
brew update                  # Update Homebrew's package list
brew upgrade                 # Upgrade all outdated packages

🔗 HomeBrew-Official

Ollama - Local and OpenSource LLMs

This is an incredibly useful tool that I find myself using constantly. Local models are fantastic for answering general or timeless questions where you need accuracy without requiring up-to-the-minute data. They’re fast, private, work offline, and have another big advantage: 🌱 a reduced environmental impact.

Instructions:

  1. Install Ollama by running the official script in your terminal. If you prefer, you can also download it manually from the official site.
    curl -fsSL https://ollama.com/install.sh | sh
  2. Browse the Model Library to find a model. Keep an eye on the size, as some models can be several gigabytes.
  3. Once you’ve picked a model, pull it using the pull command. For example, to get the Llama 3 model:
    ollama pull llama3
  4. After the download is complete, you can run the model to start a chat session. To exit, just type /bye.
    ollama run llama3

That’s it! Now, anytime you have a question, you can just run ollama run <model_name> to interact with your local LLM.

CommandDescription
ollama serveStarts the Ollama server (for API access).
ollama create <model>Creates a new model from a Modelfile.
ollama show <model>Displays details about a model.
ollama run <model>Runs a model for interactive chat.
ollama pull <model>Downloads a model from the library.
ollama listLists all downloaded models.
ollama psShows currently running models.
ollama stop <model>Stops a running model.
ollama rm <model>Removes a model from your system.

Here is an example of how you can download and run another model in one line:

ollama pull mistral && ollama run mistral

🔗 Ollama

Gemini-cli

Gemini CLI is an open-source tool that brings the power of Google’s Gemini models directly into your terminal. It provides lightweight, direct access to the API, making it a versatile utility for a wide range of tasks, from coding and content generation to problem-solving and research.

Installation

  1. Ensure you have Node.js (version 18+). You can check with node -v. If you don’t have it, install it with Homebrew:
    brew install node
  2. Install the Gemini CLI globally using npm:
    npm install -g @google/gemini-cli
  3. Run the setup and authentication command:
    gemini
    Follow the prompts to choose a theme and log in with your Google account.

Troubleshooting Tip: On some systems, the authentication flow in the terminal might not complete. If this happens, run gemini again, and then check your default web browser for the Google authentication page to complete the login.

⚠️ Unlike Ollama, only the tool is open-source, not the model. Your prompts and data will be processed by Google’s servers.

⚠️ The free tier has a rate limit (e.g., 60 requests per minute). You can upgrade to a paid plan to overcome this.

Common Commands

  • /chat save <name>: Saves your current chat with a memorable name.
  • /chat list: Lists all your saved chats.
  • /chat resume <name>: Resumes a saved chat session.
  • /compress: Replaces the current chat context with a summary to save tokens.
  • /auth: Manages your authentication settings.
  • /quit or /exit: Exits the tool.

Switching to Shell Mode

You can input ! to toggle shell mode. For example, type !pwd to see your current directory. The theme will change to indicate you’re in shell mode. To return to the chat, just input ! again.

Providing Context

Use the @ symbol to provide file or directory context for your prompts.

  • @path/to/your/file.txt Explain this text file.
  • @src/my_project/ Summarize the code in this directory.

You can find more commands in the official documentation.

Gemini CLI-Github repo

DuckDuckGo

Sometimes you just need to verify things quickly with a search engine. ddgr brings the DuckDuckGo search engine to your CLI, delivering fast results.

# Installation
brew install ddgr

Common Commands

These are the commands you’ll likely use most often:

CommandDescription
ddgr <query>Perform a search.
ddgr -n <num> <query>Limit the number of search results.
ddgr -j <query>Open the first result directly in the browser.
ddgr -w <site> <query>Restrict the search to a specific site.
ddgr -s <region> <query>Set the search region (e.g., us-en).
ddgr -x <query>Display URLs only, without opening a browser.
ddgr -C <query>Colorize the output for easier reading.
ddgr -lList your search history.
ddgr -cClear your search history.
ddgr --disable-safeDisable safe search filtering.
ddgr -hShow the help menu.

When results are displayed, you can use these keys to navigate:

Shortcut / ActionWhat it does
[number]Open the result with that number in your browser.
n or NShow the next page of results.
p or PShow the previous page of results.
o [numbers]Open multiple results (e.g., o 1 3 5).
qQuit ddgr.

Here’s an example I use extensively:

# Shows 5 results from GitHub for "awesome shell scripts"
ddgr -n 5 -w github.com "awesome shell scripts"

Profile Configuration

You can customize ddgr’s default behavior by editing the ~/.ddgrrc file (vi ~/.ddgrrc). Here are some of the options you can configure:

OptionDescription
-n <number>Set the default number of search results per page.
-CEnable colorized output.
--disable-safeDisable safe search.
-xShow URLs only by default.
-r <browser>Set the browser for opening results (e.g., firefox).
-s <region>Set the default search region (e.g., us-en).
-w <site>Restrict searches to a specific site by default.
--jsonOutput results in JSON format (useful for scripting).

🔗 ddgr-Repo

Git CLI

brew install git

I’m sure you’re familiar with the basic git commands, but you can create powerful aliases to make your workflow faster. For example, you can run git st instead of git status.

  1. Find your global git config file by running git config --list --show-origin.
  2. Open that file (e.g., vi ~/.gitconfig) and add your aliases.

Here is an example ~/.gitconfig file:

[credential]
    helper = store
[user]
    name = Eganathan R
    email = md-email@gmail.com
[filter "lfs"]
    smudge = git-lfs smudge -- %f
    process = git-lfs filter-process
    required = true
    clean = git-lfs clean -- %f
[init]
    defaultBranch = master
[http]
    proxy = http://127.0.0.1:xxxx
[alias]
    st = status
    co = checkout
    br = branch
    ci = commit
    lg = log --oneline --graph --decorate # Pretty and compact log view

I’m certain this will help you. There are many other interesting command-line tools out there (like starship, podman, tmux), but I consider them more for customization.

Thanks for reading, and have a great day ☺️