Skip to content
IFAI

AI tools

Highlight turns every app into Cursor with its new MCP interface

Published Reach 35

Highlight just launched a universal interface for MCP servers, aiming to bring Cursor-like interactions across every app on your computer. It features screen grounding that lets you mention services in-context and write with Tab, pitching itself as what Apple Intelligence and Copilot should have been.

Highlight turns every app into Cursor with its new MCP interface
Image: x.com

Cursor users know the feeling: hitting Tab to write code in-context feels like magic, but that magic vanishes the second you switch to Slack or your email client.

Highlight wants to fix that. Founder Pim de Witte just announced what he calls the first universal interface for MCP (Model Context Protocol) servers, bringing Cursor-style interactions across every app on your computer.

He even framed it as what Apple Intelligence and Copilot could have been.

Cursor powers across your entire desktop

The interface is built around two primary interactions.

First is screen grounding. The AI understands what is currently on your display, allowing you to mention and pull in MCP services right in context.

Second is native writing. Instead of bouncing between chat boxes and your work apps to copy and paste, you can generate text directly into any application simply by tapping Tab.

Why it matters

MCP is quickly evolving from a developer-focused connectivity protocol into a universal desktop operating layer. Instead of keeping MCP trapped in IDE terminals, Highlight is betting that regular knowledge workers want that exact same Tab-to-complete workflow everywhere.

We think packaging MCP behind native-feeling system keystrokes is a very smart approach to ambient AI.

That said, cross-application screen capture and text injection always face steep hurdles around latency, UI edge cases, and OS permissions. Whether it can reliably outperform built-in solutions from Apple and Microsoft will come down to everyday stability and privacy safety.

Sources (1)

  1. [1]
    twitter.com(opens in a new tab)Originaltwitter.com

Best of AI, daily