Imagine that you write in plain English what you want to do in Google Workspace (eg. workflows) and it happens just like magic. Insert text prompt, Gemini will generate the code for you and run it immediately. A dream? No, reality, thanks to my conceptual and practical idea of how to implement AI Agents in Google Apps Scripts to leverage the V8 runtime.
Ivan Kutil has explored the concept of AI Agents in Google Apps Script, enabling Google Workspace automation via plain English descriptions. Users describe their automation needs in natural language, which is then processed by Gemini API to generate the necessary code. The generated code is then executed in your Google Apps Script project.
Ivan’s solution uses the gemini-2.0-flash-thinking-exp-01-21 model, an experimental model within Vertex AI specifically designed to reveal its ‘thinking process’, resulting in more reliable code generation. The enhanced reasoning capabilities of this model are particularly beneficial for complex automation tasks, making it a powerful tool for Google Workspace customisation.
To ensure that the agent is doing the right thing, the clever bit is you can test the execution via a dry-run, where the code created with Gemini Flash Thinking is sent to an internal ‘Tester’ agent, which uses Gemini to comment on the code and summarises it in a log. It’s important to review the script before running it, as Ivan accepts no responsibility for the results of the script. Another nice feature is the generated code is stored in the Cache, so after running a dry-run and then a run, the same version will be executed within the Cache limit (currently set to 5 minutes).
This solution, which mirrors Gemini for Workspace’s ability to generate and execute basic Python code, suggests a future where Gemini for Workspace could write and execute Apps Script code for basic tasks. This has the potential to transform how users interact with and automate their Google Workspace environments.
If you are interested in a version of Ivan’s solution that incorporates my GeminiApp library, follow this link. For additional information on Ivan’s solution including setup instructions follow the source link.
Surely, you have also experienced having a presentation in which you needed to replace old content with new. Replacing text is very simple because you just need to use the Replace function, and you can either do it in the Google Slides user interface.
The problem arises when you need to replace one image with another, for example, if your corporate logo is updated to a new graphic design or if one of your favorite cloud services updates its icons (Gmail, blink blink ;-) It’s still somewhat bearable with one presentation, but what do you do when, like me, you have thousands of Google Slides files on your Google Drive?
This post explores a clever application of the Gemini API’s multimodal capabilities, created by Ivan Kutil. His code utilizes GenAI to automatically detect outdated logos within your Google Slides presentations. The original blog post (March 2024) used the Gemini Pro Vision API. As a sign of how rapidly this area evolves, Google now recommends switching to Gemini 1.5 Flash or Gemini 1.5 Pro.
Switching to the newer APIs is very straightforward as all you need to do is search the source code for models/gemini-pro-vision and replace with models/gemini-1.5-flash or models/gemini-1.5-pro.
For Apps Script projects I lean towards Gemini 1.5 Flash as it is designed for speed. I’m also always looking to opportunities to test the GeminiApp library for Apps Script and very quickly I was able to fork Ivan’s code and use Gemini 1.5 Flash using a service account.
While GeminiApp requires some initial setup, it offers significant advantages:
Easy Model Testing: Experiment with various models quickly.
Built-in Features: Includes functionalities like exponential backoff.
Google Apps Script has a new editor, which is better, nicer, and completed ready for future new features. Today, I would like to introduce you to several dirty and secret tricks, what you can do.
If you like a keyboard shortcut you’ll love this post from Ivan Kutil! For those less keen on keyboard commands there is still plenty of other productivity tips from Ivan shared in this post.
This article will show you how you can setup, train, and predict spreadsheet data with deep-learning framework Tensorflow.js. You don’t need to call REST APIs or use other 3rd parties storage and algorithm. All your data stays in your secure Google Sheet.
How to measure latency between Google Apps Script project (with UrlFetchApp) and Google Cloud Platform regions (e.g. if you are choosing region for GCP project)
