Important:
If you're signed in with a school account and are under 18, you can’t use any AI features in Workspace Studio. If an agent containing AI steps is shared with you, those steps will be removed when you open the agent.
To quickly create an agent, describe the tasks you want to automate and let Gemini build it for you.
For the best results, be specific in your description. Make sure to include:
- When the agent should start
- Which apps to use
- What steps the agent should take
To create an agent with Gemini AI:
- On your computer, go to studio.workspace.google.com.
- In the "Describe a task for Gemini" box, enter a detailed description of what you want the agent to do.
- Good example: "Every morning at 8 AM, check my unread emails. For any emails from [email protected], send a summary to my 'Urgent' space in Google Chat."
- Bad example: "Handle my morning emails."
- Click Create.
- Gemini generates the steps for your agent.
- Review each step to make sure it's correct.
- You can edit, reorder, delete, or add new steps.
- Gemini generates the steps for your agent.
- If a step needs info from a previous step, such as an email's subject line, add a variable. Learn how to use variables to pass data in your agent.
- Optional: Before you turn on your agent, to make sure it works properly, click Test run
Start.
- This runs the agent immediately so you can verify the output.
- If everything works fine, you get a “Run Completed” message.
- After you complete your edits, click Turn on.
Start.