Working with agents should feel like working with a colleague. You should be able “speak to” them not just with text chats, but by gesturing at a screen together, talking live, etc.
Anysphere co-founder Michael Truell launches Design Mode for Cursor, enabling UI editing through voice, drawing, and live gesturing
Story Overview
Cursor's Design Mode adds a visual layer to the AI agent's workflow inside its built-in browser, letting developers select rendered UI elements and issue changes that propagate back to the source code rather than staying limited to chat prompts.
A running local server remains required for the live preview
The feature loads the app via a standard dev command so the editor can inspect and alter DOM elements, component props, and visual styles in real time, yet it does not remove the need for an active local environment.
Voice and drawing inputs lack confirmation in primary sources
Posts circulating around the launch describe pointing, drawing, and speaking as interaction options, but Cursor's December 2025 announcement and subsequent changelog entries only document click-to-select plus text prompts, leaving the broader multimodal claims unverified.
Many users praised Cursor AI's new Design Mode for enabling intuitive point-draw-talk UI edits that make AI feel like a real colleague, while some objected to AI tools in general or expressed disappointment over unrelated missing features.
Most Activity
@mntruell 💯
Working with agents should feel like working with a colleague. You should be able “speak to” them not just with text chats, but by gesturing at a screen together, talking live, etc.
designing in code is now as easy as: click, chat, hold shift to multi-select.
works best with Composer 2.5
Working with agents should feel like working with a colleague. You should be able “speak to” them not just with text chats, but by gesturing at a screen together, talking live, etc.

@elonmusk @mntruell Grok Agents

@mntruell Try Grok Agents

@elonmusk @mntruell Are you bullish on the $SPCX IPO @elonmusk
I know I am - cant wait for it!

@ericzakariasson can I build cursor with cursor? pt. 2

@elonmusk @mntruell 78K7QEC9HXgUoT7U19LRKHSJaZt6eMPDGMmxJJHNpump

@mntruell

@elonmusk @mntruell What do you think about QNX

@mntruell @elonmusk 👍

@mntruell @elonmusk Moving past just typing text into a box makes the whole process feel way more natural.

@elonmusk @mntruell The xAI Cursor partnership is turning out to be amazing. Things are only accelerating!

@mntruell @elonmusk We're still using AI like it's email.
The future probably looks more like Zoom.

@elonmusk @mntruell 💯

@mntruell What version of cursor is this? Mine looks nothing like this nor does design mode work like this. 🙃

@HarshitKhemani yes

That points toward a much more natural “co-working” model for AI agents 🤝
Instead of treating agents like tools you prompt, it’s closer to a shared workspace where you can point, highlight, ask in real time, and correct instantly—almost like screen-sharing with a teammate. That would reduce friction a lot compared to back-and-forth text and make complex tasks (design, coding, planning) feel more fluid and collaborative.
The big challenge is making that interaction precise and reliable—so the agent understands gestures, context on-screen, and live conversation without misinterpreting intent.

@mntruell This is why I'd love to see a duplex voice mode in Cursor. Back and forth conversation with an orchestrator that manages agents under the hood.

@mntruell This is exactly how AI collaboration should feel — like a true colleague sitting next to you 🔥 Point, draw, talk… no more fighting with prompts. Cursor is cooking something special. Great work team!