I turned Gemini Gems into automated Google Workspace agents — here's how

Gemini
(Image credit: Shutterstock)

If you’re deep into the Google ecosystem like I am, you probably already know about Gems — Google’s custom AI chatbots, similar to OpenAI’s custom GPTs. I’ve been using them constantly because they’re great at locking in a specific tone, persona or workflow so I don’t have to keep retyping giant prompts every single time I want help with something.

However, I recently realized that even with custom Gems, I was still manually opening Gemini, typing requests and waiting for responses. It still felt like I was doing too much of the work. That’s when I started experimenting with Google Workspace Studio, Google’s no-code automation platform, and honestly, it changed the way I use AI during the workday.

Instead of chatting with my Gem directly, I can now plug it into automated workflows that run in the background. That means my AI assistant can summarize emails, analyze files, organize spreadsheets or draft replies without me constantly babysitting the process.

Latest Videos From

Once I got it set up, it genuinely felt like I had built my own behind-the-scenes, 24/7 assistant.

Whether you want to automate repetitive admin work, create smarter email workflows or build an AI-powered content pipeline, here’s exactly how I started using custom Gems inside Google Workspace Studio.

What I needed before getting started

Google Workspace screenshot

(Image credit: Future)

Before I could connect everything, there were a couple of requirements I had to figure out first.

  • The Google Drive rule: I learned pretty quickly that my custom Gem had to be grounded using Google Drive files like Docs, Sheets or PDFs. If the Gem relied on outside extensions like YouTube or Google Maps, it wouldn’t appear inside Workspace Studio.
  • Admin permissions: Workspace Studio is more of an enterprise-level tool, so, if you're using Gems on your work computer, Gemini actions need to be enabled by your organization’s Google Workspace admin before everything can work properly. If you use Workspace Studio on your own, than you are the admin and don't have to worry about this.

If you want to see if your current account supports Workspace Studio, log into your computer and go to: studio.workspace.google.com. If it lets you in, you can can start using Gemini to build automations (like telling AI to auto-summarize unread emails, or automatically saving specific Gmail attachments to a Drive folder).

If it blocks you or shows an error, it means you are either logged into a free personal Google account, or you are on a corporate Workspace account where your company's IT administrator hasn't enabled Gemini features yet.

Once you're in, you're ready to go.

How I added a custom Gem to an automated Flow

  • Sign in and click: "Create New Flow" to build a workflow from scratch.
  • I used a template. Every Flow needs something that kicks off the automation. For this, I just used the templates Google already offers. They are pretty good, but if you prefer to set your own, simply type your rule in the chat.
  • Use cases. I used mine for Gmail, but you could set a flow for Google Drive uploads, form submissions, spreadsheet updates, calendar events or really anything when it comes to "if this happens..." part of the workflow. (i.e. if I get a form submission, add it to Google Sheets).
  • I added the 'Ask a Gem' action. Next, I clicked the blue + Add step button underneath the trigger. Under the AI actions section, I selected Ask a Gem. This is where the magic happens because it lets your custom Gem process information automatically in the background. Then, connect your custom Gem.

Once you have everything set, you're going to want to test everything. Luckily, Google lets you click "Test run" to make sure the Gem handled the information correctly. If it doesn't, you can ask Gemini for help. You can ask within the space, or if you feel more comfortable, take a screenshot and use the app. If you're like me, sometimes it helps to seperate into a new app just to avoid confusion.

When my Flow worked, I switched the Flow to "Turn on" and let it run automatically.

For me, the coolest part was seeing how the Gem responds because that response then becomes usable data for the rest of the workflow. In other words, instead of of stopping after the AI does the job, I could then chain additional actions afterward.

For example, I used my Gem’s output to:

  • Gmail: Automatically create draft replies I could quickly review and send
  • Google Docs: Generate organized summaries and reports inside templates
  • Google Sheets: Route parsed data and AI analysis directly into tracking spreadsheets

Honestly, this is where it felt both too easy and too good to be true. Gemini became the admin assistant I've always needed. And even better, it is so easy to set up. Once you log into the Workspace Studio, the interface is clean and simple enough that setting up a Google Gem almost feels intuitive.

And Gemini is just a click away, so you can always ask it questions or tell it what you're trying to do. If you turn on Gemini Live, you can even share your screen.

Bottom line: I should have done this sooner

Gemini 3.5 Flash makes the entire user experience feels effortlessly fast and seamless. That's why it's a great time to explore what Gemini can do outside of the chat window. And trust me, if you can prompt Gemini, you can make the most of AI in Google Workspace Studio for autonomous productivity.

Once you get a taste of what AI can help you work with, I have no doubt that you'll increasingly start teaching the AI systems how to handle recurring tasks for you automatically.

And after trying it myself, I don’t think most people realize how close we already are to having fully personalized AI workflows — not to replace us, but assist us so we can focus on the more important stuff.


Click to follow Tom's Guide on Google News

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Subscribe to Tom's Guide on YouTube and follow us on TikTok.


More from Tom’s Guide

TOPICS
CATEGORIES
Amanda Caswell
AI Editor

Amanda Caswell is the AI Editor at Tom's Guide and one of today’s leading voices in AI and technology.

A celebrated contributor to various news outlets, her sharp insights and relatable storytelling have earned her a loyal readership. Amanda’s work has been recognized with prestigious honors, including outstanding contribution to media.

Known for her ability to bring clarity to even the most complex topics, Amanda seamlessly blends innovation and creativity, inspiring readers to embrace the power of AI and emerging technologies.

As a certified prompt engineer, she continues to push the boundaries of how humans and AI can work together.

Beyond her journalism career, Amanda is a long-distance runner and mom of three. She lives in New Jersey.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.