You've Been Using AI the Hard Way (Use This Instead)

Introduction

In this tutorial, we will explore how to enhance your productivity by leveraging AI tools directly in the terminal instead of through a traditional web browser. Utilizing tools like Gemini CLI, Claude Code, and OpenCode can drastically speed up your workflow, allowing you to work more efficiently with multiple AIs. We will cover installation, setup, and practical tips for using these tools effectively across different operating systems.

Step 1: Setting Up Your Terminal AI Tools

To get started, you need to install the necessary AI tools on your machine. Follow these steps for installation based on your operating system.

For Mac:

1. Open Terminal.
2. Install Homebrew if you haven’t already by running:

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

3. Install Gemini CLI:

   brew install gemini-cli
   

4. Install Claude Code:
   • Visit Claude Code for installation instructions.
5. Install OpenCode:
   • Visit OpenCode for installation instructions.

For Windows:

1. Use Windows Subsystem for Linux (WSL) or Git Bash.
2. Install Git Bash if not already installed.
3. Open Git Bash and install Gemini CLI:

   npm install -g gemini-cli
   

4. Follow the respective links for Claude Code and OpenCode installation.

For Linux:

1. Open your terminal.
2. Install Gemini CLI using:

   sudo apt install gemini-cli
   

3. Follow the respective links for Claude Code and OpenCode installation.

Step 2: Creating Project Aware Context Files

Once your tools are installed, you can begin creating context files that help manage projects and maintain consistency across different AI applications.

1. Create a new directory for your project:

   mkdir my-ai-project
   cd my-ai-project
   

2. Create a context file for each AI tool:
   • For Gemini:

     touch gemini_context.txt
     

   • For Claude Code:

     touch claude_context.txt
     

   • For OpenCode:

     touch opencode_context.txt
     

3. Populate each context file with relevant information and prompts specific to your project.

Step 3: Syncing Multiple AIs

To keep your AIs in sync, follow these steps to manage interactions and outputs efficiently.

1. Use a single terminal window to run multiple AI commands.
2. Redirect outputs from one AI to another as needed. For example:

   gemini-cli < gemini_context.txt | claude-cli --input -
   

3. Save outputs into a shared context file for easy access:

   gemini-cli < gemini_context.txt >> shared_output.txt
   

Step 4: Daily Workflow Optimization

Integrate your AI tools into your daily workflow to maximize efficiency.

1. Start your day by reviewing your project context files and outputs.

2. Use terminal commands to quickly switch between AIs and manage context:

   • Use the cd command to navigate between project directories.
   • Employ aliases in your shell configuration for frequent tasks.

3. Regularly update context files based on new findings or project developments.

Step 5: Advanced Features and Best Practices

Explore advanced features of each tool to further enhance your workflow.

• Gemini CLI: Utilize its advanced querying capabilities for complex tasks.
• Claude Code: Experiment with its output styles and agent configurations for tailored responses.
• OpenCode: Take advantage of local models for faster processing and reduced latency.

Common Pitfalls to Avoid

• Forgetting to update context files can lead to confusion and inefficient AI interactions.
• Overloading commands in the terminal may slow down your workflow; keep tasks modular.

Conclusion

By following these steps, you can transform your approach to using AI tools, making them faster and more efficient. Embrace the terminal environment to unlock powerful features, streamline your workflow, and keep your projects organized. As you become more familiar with these tools, consider experimenting with additional configurations to further tailor your setup to your needs. Happy coding!