Earlier this week, @greggyb and @stephr_wong kicked off Google Cloud Live, a no fluff weekly livestream series designed around hands-on building, vibe coding, and live debugging.
For their next episode, Jack Wotherspoon and Greg are taking things further with Gemini CLI. Last week they started with the basics, but now they're going to show us how automate workflows and make Gemini CLI an active part of your development lifecycle.
They'll be diving into skills, hooks, and Plan Mode. Watch them:
- Plan and implement a new app feature directly in the terminal
- Automatically lint your code and avoid writing secrets or passwords to your codebase
- Create a skill to automate the task every developer hates: writing documentation
Like last time, the Google AI team wants to answer your questions! Here's what happens when they do:
Shoutout to @ofri-peretz for getting their questions answered! Here's the replay if you want to watch the response.
Drop your questions for their next episode in the comments and they'll answer as many as they can on the stream.
Top comments (2)
Cool to see my question made it into the last episode! For this one — I'm curious how Gemini CLI's hooks handle conflicting rules when you chain multiple linters. Like if you set up a hook to auto-lint on save, does it respect existing eslint configs or does it try to impose its own fixes? That tension between AI-suggested fixes and established project standards is where things get messy in practice.
Context persistence (GEMINI.md?) + checkpointing make this killer for complex refactors.
Current session memory limits kill multi-hour workflows.
This could replace 3 browser tabs + 2 terminals daily. 🚀