Changelog

Whether you’re in Hex, using a spreadsheet, or collaborating with your colleagues, data should be there providing useful, relevant updates. Today, we’re excited to share an easy Google Sheets integration, a Cursor plugin, and the ability to bring more real-world context into your prompts by allowing the Hex Agent to accept text and image uploads. Lastly, check out the nice new streaming UX update for the agent.

📄 Google Sheets export​

You can now export to a Google Sheet in one click from cells or schedule your app to export to a Google Sheet. This is especially awesome for finance teams who want the power of the agent in Hex, but still want to easily work in a full spreadsheet view at other times. Notifications hit your inbox or Slack with a link to the spreadsheet as well as the Hex app. 

🖼️ Hex Agent now accepts text and image file uploads​

Users can now upload and attach text and image files to their prompts. Inspired by a chart you like and want the agent to help you make something similar? Just upload a screenshot. You can even use an image to help bootstrap an entire dashboard.

🧠 Interactive charts, tables, and thinking steps directly in Claude​

Our Claude Connector is now enhanced, with a native app, interactive custom-styled charts, tables and thinking steps.

When you fire off a data question from Claude, the Hex connector will spin up an interactive interface displaying the Hex agent’s thinking. Click into the steps to see which tables or semantic models the agent is referencing, and to spot-check the underlying SQL. This lets you follow the agent’s reasoning, deepen your own understanding, and point things in a new direction if needed.

Ready to dive in? Check out the Hex Connector in Claude’s directory and our documentation.

👾 Cursor Marketplace plugin​

Software engineers and vibe coders, rejoice! Hex is now a plugin on Cursor's Marketplace. You can search for Hex projects and create and interact with Threads while you code. For instructions on how to connect Hex and Cursor, check out our docs.

🕵 Two cool UX improvements for the Agent​

We also have two nice new quality-of-life improvements for those of you who love working with the Hex Agent, but hate waiting.

• Agent Streaming - This UX tweak allows you to get your agent responses faster! No more waiting for the agent to be done with every step in the analysis. Read it in real time.
• Prompt Queuing - Now, in the Notebook, you can queue up new prompts for the agent. When you enter a prompt when the agent is running, it’ll go into a queue to run right after the current prompt is done. Realized the agent is going off track, or want to nudge it down a different path? Easy. Create a whole queue of prompts. We’ll bring this capability to Threads too, coming soon!

See both of these upgrades in action below:

🕵🏼‍♀️ Introducing Context Studio​

As agents become central to analytics, data teams need a way to ensure they're responding accurately and consistently. Today, we're introducing Context Studio — a unified toolkit that brings agent observation, improvement, and deployment into one workflow. With Context Studio, teams can finally scale AI with trust and confidence.

From the Context Studio, data teams can:

• Observe agent performance across all Hex surfaces with AI-generated topics and warnings
• Diagnose issues using the Thread inspector with specific improvement suggestions
• Curate context with endorsements, semantic models, and workspace guides
• Test changes in preview before deployment with full version control
• Deploy improvements with confidence, knowing you won't break existing behavior

Context Studio is available to Admins and Managers on Team and Enterprise plans. Admins get full access to individual conversations and context management; Managers can view aggregated metrics and make context improvements.

Read our full announcement to learn more.

🤫 Sensitive mode for Threads​

You can now mark a Thread as sensitive via the data connections option. When you mark a thread as sensitive, the contents of that thread are hidden from Admins in the Context Studio.

⚡ Query mode performance and feature upgrades​

Query mode is now a first-class citizen in Hex, with significant performance improvements and feature parity with dataframe mode.

We've optimized query mode to reduce the number of queries sent to your warehouse and parallelized queries where possible, making projects that run fully against the warehouse much faster.

Query mode now includes:

• CSV downloads from query mode results
• Improved performance through query reduction and parallelization
• Easier to stay in query mode once you've opted in

Love them or hate them, dashboards are the standard for most data sharing, and where a lot of follow-up questions for data begin. Today, we’re releasing Chat with App, a way to explore your Hex apps just by talking to them. We’re also releasing some new ways to bring endorsed data into our agentic experiences to help provide consistent, relevant answers. And as requested by many of you, the launch of TV mode, so you can binge-watch your favorite apps.

💬 Chat with your published apps​

You can now chat directly with your published apps to explore and understand them in an entirely new, more intuitive way. Chat with App makes it easier for anyone to consume and navigate an app, allowing users to ask questions in plain language instead of requiring them to dig through the project's underlying logic or reach out to the data team. The agent can summarize insights, locate specific cells, adjust filters and inputs, and understand the project's underlying logic, all directly within the app experience.

Chat with App is available now for users with Explorer roles on projects that they have “Can explore” permissions to. See our documentation for more details.

Looking ahead, we’re excited to expand the agent’s capabilities to create new logic, allowing it to answer an even broader set of questions and analyses while continuing to take advantage of published app context.

🖊️ Endorsed Mode for Explorer Users​

We are introducing an Endorsed Mode for the Hex Agent in Threads. This provides a streamlined simple approach for data teams to define which data is ready for everyone to use. When in Endorsed Mode, the agent will only use project, table or semantic model context that’s endorsed, i.e. explicitly reviewed and approved by data teams. It’s on by default for Explorer users and an optional mode for Editors.

For editors, toggling on Endorsed Mode makes it easy to ask a question about a different domain that they may not be familiar with. For Explorers, this means they can trust all the data being used has been approved by their data team.

👾 Apply Endorsed Status via Hex’s API​

You can now programmatically apply endorsements to projects, semantic models, and data assets through our API. The new endpoints allow you to bulk-update statuses across your workspace, helping data teams maintain trusted context for the Hex Agent without manual clicks.

📺 TV mode​

TV Mode is designed for displaying Hex apps on monitors, TVs, or any always-on display. When in TV mode, apps automatically refresh whenever a new results from a scheduled run become available. Learn more about details and best practices for TV mode in our help docs.

🛠️ Other improvements​

• The Notebook Agent can update and add cells to a published app: The agent can now configure cells in published apps. A user can say something like "add all the 4 charts and text cells you just created to to published app" - this will use the same logic as if a user clicked the "add to app" button.
• Better query mode performance: Projects utilizing query mode now skip upstream cells that are not included in the app – the fastest query is the one you don’t need to run. As a bonus: if you choose “Query” as the return type for a SQL cell, subsequent SQL cells you create will follow suit.
• Snowflake OAuth token expiration notifications: We built an email notification for expiring personal OAuth tokens. Now, Editors whose tokens are embedded in the notebook or published app are notified prior to their token expiring to refresh it, avoiding annoying situations where a viewer can’t view an app because an Editor forgot to refresh their credentials.
• Responsive chart legend positioning: Chart legends will automatically move to the outside-bottom when the chart is very narrow (unless the legend is set to outside-top). This subtle change has the added benefit of making charts more likely to look classy on a mobile device.

📓 Notebook Agent Act II​

Thousands of teams are using the Notebook Agent every day to tap into the best of agentic capabilities right where they're already doing data work.

Today, we’re announcing new features that make the agent a leap above the rest at tackling advanced analytics: new cell types, chart styling, better context awareness, and organization tools.

The Notebook Agent can now:

• 📊 Style charts to match your preferences
• 🔢 Create input parameters and single value cells
• 🗂️ Auto-organize cells into sections
• 🗑️ Delete redundant cells and unnecessary logic
• 🪲 Debug with graph-aware context

Some of the world's most innovative companies — like Figma, Ramp, and Notion — are using the agent to supercharge analytics.

Want to get started? Check out our Notebook Agent docs.

Read our full announcement to learn more.

Data teams are playing a crucial role in governing the quality of AI responses, just like they did with trusted dashboards in the pre-AI era.

This week’s updates include tools for understanding agent behavior, a cleaner entry point into self-serve analysis on top of semantic models, and more improvements to Threads!

👀 Agent observability in the new Context Studio​

We're bringing observability and context management together in the new Context Studio — giving you the tools to manage what data is trustworthy and how agents should use it.

On Team and Enterprise plans, Admins and Managers can open up the dashboard in the Context Studio to understand how agents are being used, what users are asking, and where performance excels or needs improvement.

With this view, you can:

• Track conversation volume
• Identify your most active users
• Spot patterns in user feedback
• Filter by agent type, workspace role, time range, and more

Admins can also dive into individual conversations to:

• Review the agent's reasoning
• See what context it referenced
• Diagnose where improvements are needed

Managers get access to aggregated metrics but not individual conversations.

When you're ready to make adjustments to the agent’s behavior, you can hop over to the Context sources tab, where we’ve unified settings for workspace rules, data endorsements, and semantic projects in one place.

There is so, so, so much more coming soon here — so share your feedback and stay tuned.

🪟 Views on semantic models​

We’ve introduced views as a new layer on top of semantic models, designed for clarity and readability.

As your data models grow, you can create views in the Modeling Workbench to rename and reorganize dimensions and measures — giving business users a cleaner, more intuitive starting point for self-serve analysis.

Views can help slim down wide tables, simplify complex join trees, or even flatten relations. Now you can scale the complexity of your underlying models, then surface only curated views that are easy for end users to explore and understand.

📁 CSV uploads to Threads​

You can now upload CSVs to Threads, and the Hex Agent can craft queries against the data in the file provided. The agent is also capable of joining the CSV data with existing warehouse data, enabling cross-source insights from a simple file upload.

This is available to users with the Explorer role or higher who have permission to access non-semantic data in Threads. Admins can configure agent data access and data connection defaults in the Threads settings page.

🧵 More Threads improvements​

We’ve made a bunch of additional improvements to Threads!

You can now copy and download charts, pivots, and tables in Threads. Want to add a chart from Threads to your presentation slides? Or import the data in a table into a spreadsheet? You can copy results to your clipboard, or download them as images or CSV files.

When a chart or pivot is built directly on a semantic model, we’ll tag the name of the model directly in the footer of the visualization. This provides a standardized way to check the data source at a glance.

We’ve also made Threads faster at answering simple quick questions. When you ask a question, the agent will gauge the complexity; if it’s deemed an “easy” question, the agent will spend less time thinking and return a quicker response.

Give these improvements a spin by trying our new Threads browser shortcuts! hex.new has always been the shortcut for creating a new project in Hex, and now you can enter threads.new in your browser’s address bar to create a new thread. For an even quicker shortcut, use threads.new?prompt= followed by your question.

📐 User-level agent instructions​

If you want to steer the agent to behave in certain ways — whether it’s the SQL style or just the agent’s tone — you can now add user-level rules files to customize the styling and preferences of the agent’s responses.

From the custom instructions modal, you can also see the workspace context that Admins have added for your organization.

Other improvements​

• Better bulk column editing experience: The Columns menu in the lower right corner of SQL and table cells got a UX refresh and new capabilities! You can now reorder, hide, and pin columns from one tab, and bulk format them in a separate tab.
• Chart upgrades by the Notebook Agent: If you use the Notebook Agent to modify a legacy chart cell, the agent will automatically upgrade the chart cell on your behalf in order to make the changes you ask for.
• Start Threads from workspace search: You can use the keyboard shortcut Cmd+K to open workspace search, type in your question, and start a thread from there.