Last month, we introduced the ability to take basic actions to create and modify tables via the Google Sheets API. This week, we’re expanding the API functionality to include the ability to write and read file and people smart chips.
Following last month’s update for tables, it was recently announced that the Google Sheets API now allows developers interact with the file and people smart chips. This update provides developers with direct access to manage these types of smart chips, which were previously only available through the user interface. For detailed documentation and code snippets on implementation, refer to the official Google Developers support page.
Here is a quick overview of what you can do:
Reading Smart Chips: You can retrieve data for all smart chips within a cell. The API response will detail the properties for chipped sections of text and will also identify the plain text sections of the cell. As noted in the documentation to get the chip’s display text, you must request the formattedValue.
Adding Smart Chips: You can now insert smart chips for people and files into cells.
People Chips: These can be created to link to an individual using their email address. You can also specify the display format for the person’s name.
Rich Link Chips: These can be used to create a chip that links to a Google Drive file. Note that writing these chips requires your application to have at least one Drive OAuth scope.
In the feature image above you can hopefully see how a recruitment tracker could be created and read. In the example people chips are used for the ‘Applicant’ and ‘Interviewer’ columns, providing immediate access to contact information. A file chip is used in the ‘Resume’ column to link directly to the candidate’s document in Google Drive. A developer could automate this entire process. For instance, a script could be triggered when a new application is submitted, automatically adding a new row to the tracker complete with the relevant person and file chips, streamlining the entire recruitment workflow.
For those using Google Apps Script, these new functionalities are accessible via the Google Sheets Advanced Service. This update provides better methods for managing smart chips in Google Sheets.
As powerful as Google’s new AI tools are, those of us who live in Google Workspace have likely encountered some of their current limitations. One of the most significant for me is in Google Chat. While Gemini can analyse recent conversations, it seems to be limited to roughly the last 20 messages or 7 days of history. Crucially, it also can’t see or process messages sent by apps or webhooks, which are often the lifeblood of a technical team’s space.
This creates a knowledge gap. How can you analyse project history, debug an automated alert, or get a complete overview of a conversation if your tools only see a fraction of the data?
To solve this I’ve created an automated solution using Google Apps Script. This tool can archive an entire Google Chat space history into two powerful, synchronised formats:
A structured JSON file that serves as the definitive, machine-readable “source of truth”.
A clean, human-readable Google Doc, which is automatically generated from the JSON data.
This dual-output approach provides the best of both worlds: a portable data file perfect for backups or analysis, and a browsable document ideal for quick searches or for use as a direct data source in a Gemini Gem or other tools like NotebookLM.
Best of all, once configured, you can set up a time-driven trigger to run the script automatically, ensuring your archives are always kept up-to-date without any manual intervention.
Prerequisites
Before you begin, please ensure you have the following:
A Google Workspace Account: For setup this can only be a Google Workspace account, once it is configured it is possible to test with a personal consumer Gmail account.
Access to Google Cloud Platform (GCP): You will need the ability to create a new GCP project and enable APIs. The guide below will walk you through creating a fresh project, but if you are using a Google Workspace account, you may need to request a project depending on your organisation’s policies.
User-Based Access: This script runs as you, the user who authorises it. Therefore, it will only be able to see and archive Google Chat spaces that you are a member of.
The Setup Guide
Getting started is straightforward. The core logic is pre-packaged in a Google Sheet template. You just need to make a copy and connect it to your own Google Cloud project.
Step 1: Copy the Google Sheet Template
First, make your own personal copy of the Chat Archiver template.
Create a new project. Give it a memorable name (e.g., “My Chat Archiver”).
Enable the Google Chat API:
In your new project, navigate to the “APIs & Services” > “Enabled APIs & services” and enable the Google Chat API and Google Drive API , or Enable the APIs
If you see a message that says Google Auth platform not configured yet, click Get Started
Add an App name and support email
Choose a user type. Internal is best if you’re on a Workspace account. External will work for anyone but require you to add yourself as a test user (you can add up to 100 testers without the need of going through verification).
Click Next.
Under Contact Information, enter an Email address where you can be notified about any changes to your project, then click Next.
Under Finish, review the Google API Services User Data Policy and if you agree, select I agree to the Google API Services: User Data Policy, then click Continue.
In the Apps Script editor, go to Project Settings (the cog icon ⚙️).
Under “Google Cloud Platform (GCP) Project”, paste the Project Number and click Set Project.
Your copy of the sheet is now correctly configured and authorised to make requests to the Chat API on your behalf.
Step 3: Run Your First Archive
Refresh your copied Google Sheet. A new menu named Chat Archiver will appear.
In the Config tab, paste a Google Chat space ID or full room URL and add your email address (you can share your copy of the Google Sheet with other users who will be able to jump to this step).
Click Chat Archiver > Archive All Spaces.
The first time, a dialogue will appear asking for authorisation. Grant the permissions to allow the script to run.
The script will now run, creating both a JSON file and a Google Doc, and will place their links in the corresponding columns in the sheet.
(Optional) Click Chat Archiver > Setup 15-Minute Trigger so setup a trigger to update the archive documents every 15 minutes.
Understanding the Outputs
The script produces two distinct files, each with a specific purpose:
The JSON File: The primary output is a structured JSON file. It takes the chat messages and uses a nested format, which groups replies under their parent message, makes it easier for data analysis with LLMs.
The Google Doc: Alongside the JSON, the script generates a clean Google Doc. This format is perfect as knowledge for a Gem or as a data source in NotebookLM. Crucially, this document is completely regenerated from the JSON data on every run. This means it is a “read-only” artefact; any manual changes will be overwritten.
Important Considerations and Limitations
While this solution is powerful and robust for many use cases, it’s important to understand its design limitations and potential areas for future improvement.
Access is User-Based (By Design) – The script runs with the permissions of the user who authorises it. This means it can only access and archive Google Chat spaces that the user is currently a member of.
Future Improvement: For a true, organisation-wide backup solution, this script could be adapted to use a Service Account. With domain-wide delegation granted by a Workspace administrator, a service account could access any chat space in the domain, regardless of membership. This is a more complex setup but is the standard for enterprise-level automation.
Archives are Text-Only – This solution is focused on archiving the textual content of conversations. It does not save file attachments like images, PDFs, or videos. The JSON data will note that a message contained an attachment, but it won’t contain the file itself.
Data is Append-Only – The script is designed to add new messages to the archive. It does not synchronise edits or deletions. If a message is edited in Google Chat after it has been archived, the JSON file will still contain the original text. If a message is deleted, it will remain in the archive.
Apps Script Execution Limits – Google Apps Script has built-in limits to prevent abuse.
Execution Time: A single script run is limited to 6 minutes for standard Gmail accounts or 30 minutes for Google Workspace accounts. For an exceptionally large space (many hundreds of thousands of messages), the very first archive could potentially time out. Using the “Archive Start Date” field is the best way to manage very large initial backfills.
File Size: There is a 50 MB limit on the size of a data blob that Apps Script can create. As the JSON archive grows, a single file may eventually hit this limit. For extremely large spaces, the script could be modified to create new files on a yearly or quarterly basis.
Not a Compliance Substitute for Google Vault – For organisations that require eDiscovery, legal holds, and retention policies for compliance reasons, Google Vault is the official, authoritative tool. This script should be seen as a powerful data extraction and analysis tool, not a substitute for an official compliance solution.
Ideas for Your Archive
Now that you have your complete chat history in a structured format, what can you do with it? Here are a few ideas to get you started:
Create a Project FAQ with Gemini: Provide the generated Google Doc or JSON file to Gemini and ask it to synthesise a “Frequently Asked Questions” document about a specific project. It can pull out key decisions, find links to important resources, and identify common questions your team has answered.
Generate an Infographic Summary: Use a prompt like, “Based on this JSON data, identify the top 5 most active members, the busiest day of the week, and the main topics discussed. Present this as data for an infographic.” You can use the Gemini App ‘canvas’ to create a webpage or feed the structured output into a design tool.
Onboard New Team Members: Give a new team member the Google Doc archive and ask Gemini to act as a project expert. They can ask questions like, “What was the final decision on the ‘Phoenix’ feature?” or “Summarise the key discussions from last March,” and get instant, context-aware answers.
Perform Topic and Sentiment Analysis: The structured JSON file is perfect for programmatic analysis. You could ask Gemini to write a simple script (or use a Colab notebook) to parse the messages, perform sentiment analysis over time to track team morale, or use natural language processing to identify recurring topics and pain points.
Build a Project Timeline: Ask Gemini to scan the archive for key dates, deadlines, and project milestones mentioned in the chat, then format the results as a timeline or a table that you can use for project retrospectives and reports.
Summary
Ultimately, this tool is about providing an easy way for you to unlock the value hidden in your team’s conversations. By moving beyond the current limitations and including every message, you create a better true source of truth for analysis and reporting. While we hope future updates to Gemini for Workspace will make this kind of deep integration seamless, for now, this solution gives you more control over your data and a process you can experiment with. I hope this guide helps you get started quickly and if you would like to explore enterprise level solutions get in touch.
The landscape of what’s possible within the browser is quietly undergoing a significant shift, and for Google Workspace Add-on developers, this could be a game-changer. Chrome’s AI mission is simple yet powerful: to ‘make Chrome and the web smarter for all developers and all users.’ We’re seeing this vision begin to take shape with the emergence of experimental, built-in AI APIs in Google Chrome, designed to bring powerful capabilities using models like Gemini Nano directly to the user’s device.
There is a growing suite of these on-device APIs. This includes the versatile Prompt API, specialised Writing Assistance APIs (like Summarizer, Writer, and Rewriter), Translation APIs (Language Detector and Translator), and even a newly introduced Proofreader API. For many existing Workspace Add-on developers, some of the more task-specific APIs could offer a relatively straightforward way to integrate AI-powered enhancements.
However, my focus for this exploration, and the core of the accompanying demo Add-on being introduced here, is the Prompt API. What makes this API particularly compelling for me is its direct line to Gemini Nano, a model that runs locally, right within the user’s Chrome browser. This on-device approach means that, unlike solutions requiring calls to external third-party GenAI services, interactions can happen entirely client-side. The Prompt API provides web applications, including Google Workspace Editor Add-ons, with an open-ended way to harness this local AI for text-based generative tasks.
To put the Prompt API’s text processing abilities through its paces in a practical Workspace context, I’ve developed a Google Workspace Add-on focused on text-to-diagram generation. This post delves into this demonstration and discusses what on-device AI, through the versatile Prompt API, could mean for the future of Workspace Add-on development, including its emerging multimodal potential.
Why This Matters: New Horizons for Google Workspace Developers
Using an on-device LLM like Gemini Nano offers several key benefits for Workspace Add-on developers:
Enhanced Data Privacy & Simplified Governance:Sensitive user data doesn’t need to leave the browser, meaning no external API calls are made to third-party servers for the AI processing itself, which is a huge plus for privacy and can simplify data governance including Google Workspace Marketplace verification and Add-on data privacy policies.
Potential for Cost-Free GenAI (with caveats!): Client-side processing can reduce or eliminate server-side AI costs for certain tasks. Remember, “Nano” is smaller than its cloud counterparts, so it’s best for well-scoped features. This smaller size means developers should think carefully about their implementation, particularly around prompt design to achieve the desired accuracy, as the model’s capacity for understanding extremely broad or complex instructions without guidance will differ from larger models.
Improved User Experience & Offline Access:Expect faster interactions due to minimise network latency.
The biggest takeaway here is the opportunity to explore new avenues for GenAI capabilities in your Add-ons, albeit with the understanding that this is experimental territory and on-device models have different characteristics and capacities compared to larger, cloud-based models.
Proof of Concept: AI-Powered Text-to-Diagram Add-on
To showcase the tangible possibilities of on-device text processing, the demonstrator Add-on (available in the Nano-Prompt-AI-Demo GitHub repository) focuses on a text-to-diagram use case:
Users can describe a diagram in natural language (e.g., “flowchart for a login process”).
The Add-on then uses the Gemini Nano API via the Prompt API to convert this text description into MermaidJS diagram code.
It also allows users to directly edit the generated MermaidJS code, see a live preview, and utilise an AI-powered “Fix Diagram” feature if the code has errors.
Finally, the generated diagram can be inserted as a PNG image into their Google Workspace file.
Nano Prompt API Demo
This example illustrates how the Prompt API can be used for practical tasks within a Google Workspace environment.
Under the Bonnet: Utilising the Chrome Gemini Nano Prompt API for Text
The Add-on interacts with Gemini Nano via client-side JavaScript using the LanguageModel object in the Sidebar.html file. I should also highlight that all of the Sidebar.html code was written by the Gemini 2.5 Pro model in gemini.google.com, with my guidance which included providing the appropriate developer documentation and this explainer for the Prompt API.
The Add-on’s core logic for text-to-diagram generation includes:
Session Creation and Prompt Design for Gemini Nano:A LanguageModel session is created using LanguageModel.create().
Generating Diagrams from Text: The user’s natural language description is sent to the AI via session.prompt(textDescription).
AI-Powered Code Fixing: If the generated or manually entered MermaidJS code has errors, the faulty code along with the error message is sent back to the model for attempted correction.
Given that Gemini Nano is, as its name suggests, a smaller LLM, careful prompt design is key to achieving optimal results. In this demonstrator Add-on, for instance, the initialPrompts (system prompt) play a crucial role. It not only instructs the AI to act as a MermaidJS expert and to output onlyraw MermaidJS markdown, but it also includes two explicit examples of MermaidJS code within those instructions.
Providing such “few-shot” examples within the system prompt was found to significantly improve the reliability and accuracy of the generated diagram code from text descriptions. This technique helps guide the smaller model effectively.
Navigating Experimental Waters: Important Considerations (and Reassurances)
It’s important to reiterate that the majority of AI APIs are still experimental. Functionality can change, and specific Chrome versions and flags are often required. I recommend referring to official Chrome AI Documentation and Joining the Early Preview Program for the latest details and updates.
Before you go updating your popular production Google Workspace Add-ons developers should be aware of the current system prerequisites. As of this writing, these include:
Operating System: Windows 10 or 11; macOS 13+ (Ventura and onwards); or Linux.
Storage: At least 22 GB of free space on the volume that contains your Chrome profile is necessary for the model download.
GPU: A dedicated GPU with strictly more than 4 GB of VRAM is often a requirement for performant on-device model execution.
Currently, APIs backed by Gemini Nano do not yet support Chrome for Android, iOS, or ChromeOS. For Workspace Add-on developers, the lack of ChromeOS support is a significant consideration.
However, Google announced at I/O 2025 in the ‘Practical built-in AI with Gemini Nano in Chrome’ session that the text-only Prompt API, powered by Gemini Nano, is generally available for Chrome Extensions starting in Chrome 138. While general web page use of the Prompt API remains experimental this move hopefully signals a clear trajectory from experiment to production-ready capabilities.
Bridging the Gap: The Hybrid SDK
To address device compatibility across the ecosystem, Google has announced a Hybrid SDK. This upcoming extension to the Firebase Web SDK aims to use built-in APIs locally when available and fall back to server-side Gemini otherwise, with a developer preview planned (see https://goo.gle/hybrid-sdk-developer-preview for more information). This initiative should provide a more consistent development experience and wider reach for AI-powered features.
A Glimpse into the Future: Empowering Workspace Innovation
On-device AI opens new opportunities for privacy-centric, responsive, and cost-effective Add-on features. While the demonstrator Add-on focuses on text generation, the Prompt API and the broader suite of on-device AI tools in Chrome offer much more for developers to explore
Focusing on Unique Value for Workspace Add-ons
It’s important for developers to consider how these on-device AI capabilities—be it advanced text processing or new multimodal interactions which support audio and image inputs from Chrome 138 Canary—can be used to extend and enhance user experience in novel ways, rather than replicating core Gemini for Google Workspace features. The power lies in creating unique, value-added functionalities that complement native Workspace features.
Explore, Experiment, and Provide Feedback!
This journey into on-device AI is a collaborative one and Google Workspace developers have an opportunity to help shape on-device AI.
Try It Out: Follow setup instructions to experience on-device AI with the demo, and consider exploring multimodal capabilities for your own projects by referring to the latest Early Preview Program updates.
Provide Feedback: Share your experiences either about the example add-on or through the Early Preview Program.
I hope you have as much fun working with these APIs as I have and look forward to hearing how you get on. Happy Scripting!
Following the improvements made to tables in Google Sheets in March and April, we’re excited to introduce API support for tables. Now, users will be able to take basic actions to create and modify tables via the Sheets API.
Following enhancements to Google Sheets tables last year, Google recently highlighted an update for developers: the Sheets API now supports tables. This is great news for the Google Workspace developer community, as it allows for direct programmatic creation and modification of tables, moving beyond previous reliance on workarounds.
For a while, developers have found clever ways to interact with table-like structures, such as Kanshi Tanaike’s notable 2024 solution using a mix of Apps Script and Sheet formulas. While these methods were very clever, the new, direct API support offers a more robust and straightforward way for interacting with tables in Google Sheets. For Google Apps Script users, for now this will require using the Google Sheets Advanced Service to call these new table methods, as direct integration into the SpreadsheetApp service hasn’t been announced at this time.
Key API Capabilities for Tables:
The Sheets API now lets developers:
Add Tables: Create tables with defined names, ranges, and specific column properties (like ‘PERCENT‘ or ‘DROPDOWN‘ with validation).
Update Tables: Modify table size (add/remove rows/columns), and toggle table footers. The API also provides methods like InsertRangeRequest and DeleteRangeRequest for more granular control.
Append Values: Easily add new rows to the end of a table using AppendCellsRequest, which intelligently handles existing data and footers.
Delete Tables: Remove entire tables and their content (DeleteTableRequest) or just the formatting while retaining data (DeleteBandingRequest).
Utilise Column Types: Work with various column types including numeric, date, dropdown, smart chip, and checkbox.
This is a sample Google Apps Script designed to replace all instances of specific text within a Google Slides presentation, while simultaneously applying a desired text style. The built-in Presentation.replaceAllText() method within the Google Slides service is limited; it efficiently replaces text strings but lacks the functionality to modify text formatting during the replacement process. This limitation poses a challenge when aiming for styled text replacements. This report presents a detailed script solution that overcomes this constraint.
Have you ever needed to replace text across your Google Slides presentation but also wanted to apply specific formatting to the new text at the same time? The standard replaceAllText() method in Apps Script is handy for bulk text replacement, but it falls short when you need to control the styling – like font, size, or colour – during the replacement process.
Community contributor Kanshi Tanaike has developed a clever solution to overcome this limitation. Tanaike has shared a Google Apps Script function that not only finds and replaces text throughout all elements in your slides (including shapes, tables, and grouped objects) but also applies your desired text styles simultaneously.
The script works by iterating through the elements on each slide. When it finds the text you want to replace, it uses TextRange methods to perform the replacement and apply the specified formatting attributes, such as font family, size, colour, bold, italics, underline, and more.
This approach provides significantly more control than the built-in method, allowing you to ensure that automatically replaced text matches the exact styling you need for visually consistent and polished presentations. Tanaike’s post includes the full script, configuration details for specifying the text and styles, and sample slides showing the results.
Gemini API now generates images via Flash Experimental and Imagen 3. This report introduces image evolution within conversations using Gemini API with Google Apps Script.
The Gemini API recently gained the ability to generate images. Taking this a step further, Kanshi Tanaike has explored how to create evolving images within a conversation using Google Apps Script.
Often, you might want to generate an image and then iteratively add or modify elements in subsequent steps. Kanshi’s approach cleverly uses the chat functionality of the Gemini API (gemini-2.0-flash-exp model). By sending prompts sequentially within a chat, the API uses the conversation history, allowing each new image to build upon the previous one. This enables the generation of images that evolve step-by-step based on your prompts, as demonstrated in the original post with examples like drawing successive items on a whiteboard.
This technique is particularly useful because, as noted in the post, using chat history provides better results for this kind of sequential image generation compared to generating images from isolated prompts.
Kanshi Tanaike’s original post includes a detailed explanation, setup instructions (including API key usage and library installation ), and complete sample code snippets that you can adapt for your own Google Workspace projects.
I’ve created a small (pseudo) class that more easily clears and creates conditional formatting rules in a Google Sheet tab with Google Apps Script. Why? Well in Google Apps Script, conditional formatting rules are all or nothing. You can only ever set, get or clear ALL rules in a specified Google Sheet tab.
Conditional formatting in Google Sheets can be a powerful tool. However, Google Apps Script’s native setConditionalFormatRules() method replaces the entire set of rules, requiring the retrieval and reconstruction of the complete rule array to add, modify, or remove a single rule. This can be cumbersome and inefficient.
Fortunately Scott Donald has developed Range_ConditionalFormatting(), a valuable utility that simplifies conditional formatting management. This includes a SetRule method that allows you to add new rules without overwriting existing ones, and the position parameter provides precise control over rule order, which is crucial as conditional formatting rules are applied sequentially.
Scott’s post includes all the code you need to get this working in your own project along with a number of example snippets. This makes managing conditional formatting much more efficient and manageable.
Today I had the task of implementing a newsletter for internal updates and decided to use Google Apps Script since I was already using it for managing the project data. Based upon past experiences with emails using HTML I need this area was full of landmines and I didn’t want to navigate through it. 💣
So I decided to use MJML with Google Apps Script.
Sending emails using Google Apps Script is a common task, however, creating HTML emails that render consistently across different email clients and that look visually appealing can be challenging.
Justin Poehnelt’s post on DEV Community, “Using MJML in Google Apps Script to Send Beautiful Emails,” offers a solution to this problem. MJML is a markup language that makes it easy to create responsive HTML emails. It removes some of the complexities of HTML email development, allowing users to focus on the content of their emails.
In the post, Justin explains how to use MJML with Google Apps Script to send beautiful, responsive emails. He also mentions that he has published a library, MJMLApp, that “hides the gnarly bits” of using MJML with Google Apps Script. This library is available on GitHub in the mjml-apps-script repo where you can find the library ID and basic usage information.
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.
If you have domain-specific knowledge that you want an LLM to leverage, you probably have a use case for fine-tuning. Fine-tuning can significantly improve how well the model understands and responds to your queries, whether it’s legal documents, medical texts, financial reports, or niche industry data.
The most crucial step in this process is structuring your data correctly. If your dataset is well-organized and formatted properly, the rest of the workflow becomes much more manageable. From there, it’s just a matter of setting up a few configurations and automating parts of the process with Apps Script. That’s where things get interesting and surprisingly efficient.
