Gemini CLI: Free & Open Source AI Coding Agent

✨ Join the Ship My App cohort: https://mrc.fm/cohort 👈 Join the live stream to explore Google's brand new, free, and open source Gemini CLI. This hands on tutorial will cover installation and show you how to leverage the powerful Gemini 2.5 Pro AI agent directly in your terminal. We'll live build a project to test its limits, including integrating with Model Context Protocol (MCP) and generating images with the new Imagen 4 API. 0:00 Intro 0:25 What is Gemini CLI? 1:47 Installation & Setup 4:48 First Attempt: Building an Imagen 4 App 7:38 Showcase: Google's Imagen 4 9:23 The First App is Running (but it's ugly) 14:34 The Process is Stuck 17:34 Second Attempt: Starting Over 21:30 Third Attempt: It's Painful 36:20 Fixing the Hallucinated API Call 46:07 Giving Up on Imagen 4, Pivoting to Replicate 51:33 Success! The Replicate App Works! 56:03 ...And I Broke It Again 1:00:11 Gemini CLI vs Claude Code 1:08:14 Gemini Finally Fixes the App 1:13:10 Testing the Model Context Protocol (MCP) 1:26:47 Using MCP to Create a Supabase Database 1:40:46 Success! The MCP Works! 1:42:36 Creating an Edge Function with MCP 1:56:29 Final Thoughts & Conclusion