AI Agents +
Applied Training.

New

• Turn on Memory! Hatz AI now remembers helpful details about you across conversations. As you chat, the AI saves durable context and uses it to personalize future responses without you having to repeat yourself. Memories are per-user and never shared with others in your organization

• You can now restrict an agent's OneDrive access to a specific set of folders and/or files, rather than granting access to your entire OneDrive. Scoping supports individual files, folders, or a mix of both in a single configuration. Item names in the UI stay current when content is renamed. OneDrive authentication has also been updated to use a popup sign-in flow, so you stay on the same page while connecting your Microsoft account.

Improved

• API Documentation has been updated to include details for disable LLM routes and workflow data as well a new endpoint to return and update packages with richer data.

New

• The AI Preferences page now includes a default LLM selector — choose your preferred model once and it will automatically load at the start of every new chat session, no manual switching needed.

• New Stripe Integration is now available in Chat, Agents, and Workflows. Connect Stripe to give Hatz live visibility into your billing data so you can look up customers, manage subscriptions, review invoices, track payments, and take action on your Stripe account directly from chat.

• New LLM GPT‑5.4 added to the model selector for chat, apps, agents, and workflows.

Photo provided by Open AI

Improved

• Usage data in the dashboard now updates in near-real time, capped at the most recent 15-minute boundary.
• Improved run reliability for Workflows via the Hatz Workshop Assistant.
• Chat no longer autoscrolls as text is being generated

New

New LLM Gemini 3.1 Flash Image (Nano Banana 2) added to the model selector for chat.

Nano Banana 2 brings the high-speed intelligence of Gemini Flash to visual generation, making rapid edits and iteration possible. It makes once-exclusive Pro features accessible to a wider audience, including:

Advanced world knowledge: The model pulls from Gemini’s real-world knowledge base, and is powered by real-time information and images from web search to more accurately render specific subjects. This deep understanding also helps you create infographics, turn notes into diagrams and generate data visualizations.

Precision text rendering and translation: Nano Banana 2 allows you to generate accurate, legible text for marketing mockups or greeting cards. You can even translate and localize text within an image to share your ideas globally.

-Google

New

• The agent editor has a new right panel: the Build Assistant that helps you write and refine your agent's instructions as you build. Quick-action buttons let you generate instructions from scratch, improve what you've already written, or change the tone and style — all without leaving the editor. AI suggestions appear inline with an "Apply to Instructions" button, so you can review and confirm before anything changes.

• Customize the Greeting Message in the agent editor to set the opening message an agent uses to start every conversation.

• Agents can now override personalization. When building an agent, users can toggle off personalization to keep the agent performance standard across users.

• The Usage Analytics API is now available and fully documented. Admins can query credit consumption, user activity, and more via the API.

Improved

• Notable enhancements to the speed and performance of the Hatz Workshop Assistant.
• Faster loading times for Dashboard analytics.

Fixed

• The OpenAI Responses Endpoint now supports standard Bearer token authentication, with updated API documentation to match.
• Fixed an error that prevented LMS Report Generation from completing successfully.
• Resolved an issue causing Dashboard load failures for tenants with large usage volumes and user bases.
• Fixed an error preventing successful Document Processing Step outputs in Workflow Runs.

New

• Personalization in AI Preferences: a user-level control center to manage how Hatz responds across all chats and models. Set your details (name, job, timezone), choose default style/tone/response length, add custom instructions, and toggle personalization to shape reply behavior without changing model capabilities.

• Chat Folders — create up to 15 personal, color‑coded folders to keep your conversations beautifully organized and easy to find. Organize your conversations exactly how you work, and pin your most important folders to the top for instant access.

• Agent Tool Visibility - See which tools an Agent is using in real time, directly in chat. If a connected integration in an Agent needs to be re-authenticated, you’ll get an in-chat alert with a prompt to reconnect—and guide you to fix it without ever leaving the chat.

• Automatic Tool Selection is now available when building Agents. Toggle this on during Agent setup to allow your agent to intelligently pick the right tools for the job on its own, no manual configuration needed.

• Workshop Items in an End Customers tenant created by their MSP now display the MSP tenant name as the creator.
• Hatz now offers an OpenAI‑compatible API endpoint. If a product supports a custom OpenAI base URL (like many coding tools/agents), customers can point it at Hatz.
• Added CSV export functionality to user management tables. Users can now download their user data (including name, email, role, last login date, and credits used) as a CSV file via a new download button.
• New Airtable integration is now available with improved functionality and reliability. Legacy integration sunsets March 15, 2026.
• New LLM Gemini 3.1 Pro Preview added to the model selector for chat, apps, agents, and workflows.

-Photo provided by Google

Improved

• The Integrations tab has been renamed to the Connections tab. This updated view makes it easier to search, filter, and sort available Hatz integrations, while providing a cleaner, easier-to-follow view of connection status.

• Tenant Management capabilities have been enhanced with new features and updates.