Native In-App AI Assistant with Interactive UI for End-Users

Feature Requests
1

Native In-App AI Assistant with Interactive UI for End-Users

What’s the feature?

A native, conversational AI widget (supporting text and voice) embedded directly into published Teable applications, designed specifically for end-users rather than database builders.
Instead of generating plain-text responses or forcing users to navigate complex menus and filter grids, this assistant brings the app’s interface directly into the chat. Key capabilities include:
Interactive UI in Chat: The AI can render actual Teable elements—such as Record Cards, Data Entry Forms, Charts, and Action Buttons—right inside the conversation thread based on the user’s prompt.
Strict Permission Adherence (RBAC): The agent operates strictly within the boundaries of the logged-in user. It must 100% respect Teable’s existing Role-Based Access Control and Row/Column-level security, ensuring users can only retrieve data or trigger actions they are explicitly authorized for.
Multi-Platform: Fully functional on both web and mobile app views to support on-the-go workflows.

Your use case

Imagine you build a custom CRM or Project Tracker for a non-technical client or team. Currently, to get their work done—especially on mobile—they have to navigate sidebars, switch between different views, and manually apply filters. This often leads to low adoption rates because the interface feels like a complex database rather than an intuitive app.
With a native In-App AI Assistant, a sales rep in the field could simply tap the microphone icon and ask: “Show my sales performance for this month, and who I need to follow up with today.”
Instantly, right inside the chat thread, the AI responds by surfacing two specific UI elements:
A mini Chart/Graph summarizing their monthly target progress.
The specific Teable Record Cards for the three clients they need to call, complete with an interactive “Update Status” action button directly on each card.
The user clicks the button, updates the record, and closes the app without ever opening a grid view, hunting for a menu, or seeing irrelevant data. This bridges the gap between powerful backend databases and a frictionless, voice-driven frontend that non-technical users will actually love using.

1 Like
2

Hi @Ahmed_AegHub ,

Thanks for the detailed suggestion. We understand what you’re looking for. This would involve significant work around app integration, interactive UI, and permission management.

We appreciate the feedback and will keep it in mind for future consideration.

3

Hi @Ahmed_AegHub, thanks for the detailed write-up. The sales-rep example makes the need very clear.

A good part of this already works with Teable itself, without embedding a separate assistant in each published app:

Invite the client’s team as collaborators and give them roles in the authority matrix, so each person only sees the records and fields they’re allowed to.
They talk to Cuppy directly, on the web today and in our iOS and Android app, which is launching soon (sneak peek). The app has voice input, so a rep in the field can just ask “who do I need to follow up with today?”
Cuppy stays within each user’s permissions, including the authority matrix’s record- and field-level rules. If a rep’s role only covers their own deals, that’s all Cuppy will read or change for them.
Answers go beyond text. Cuppy can show artifacts and images right in the conversation, such as a chart of this month’s progress, ask follow-up questions with tappable options, and link to the records it updates.