
Superhuman Go brings your favorite agents to the cursor in every app and tab. Highlight anything to get started.
The Hook
“(No spoken dialogue — High-energy audio transition/riser leading to a sub-bass drop/impact.)”
Product Reveal
“(No spoken dialogue — Driving electronic beat with pizzicato synths and deep sub-bass.)”
Feature Teaser
“(No spoken dialogue — Sub-bass drop/impact with a subtle sweep sound, followed by the driving beat.)”
Call to Action
“(No spoken dialogue — Driving electronic beat continues, fading out slightly towards the end.)”
Synthesizable in Remotion · key components of the contextual-ai-workflow template.
A cursor smoothly highlights a section of text, followed by a quick, responsive pop-up of a contextual action menu directly at the cursor's position.
Use `interpolate` for smooth cursor movement and text background color change. Implement a `spring` animation for the menu pop-up, with a slight overshoot for responsiveness. Utilize `opacity` and `scale` transitions for the menu items.
A compact menu expands from a single input field into a list of actionable options, with each option appearing sequentially with a subtle fade and slide-up effect.
Employ `sequence` to stagger the appearance of menu items. Each item should use `spring` for its `y` position and `opacity`, creating a cascading reveal. The initial input field can use a `scale` and `opacity` transition for its transformation into the expanded menu.
A loading state with animated dots transitions into a concise, well-formatted text explanation within a modal window, which then can be dismissed.
Use `loop` for the loading dots animation. The explanation text should appear with a `typewriter` effect or a rapid `fade-in`. The modal itself should use a `spring` animation for its `scale` and `opacity` when appearing and disappearing, ensuring a clean, snappy feel.
Purpose. Introduce the core interaction: highlighting text to activate the agent.
Execution. Cursor highlights text, immediately followed by a small, responsive UI pop-up.
Purpose. Showcase the range of actions available through the agent.
Execution. The initial UI expands into a clear menu of options like 'Explain', 'Cite', 'Scan for AI'.
Purpose. Demonstrate the agent's capability to provide instant, relevant information.
Execution. A loading state transitions into a concise, accurate explanation within a clean modal.
Purpose. Emphasize the product's non-intrusive integration into the user's existing workflow.
Execution. The explanation modal is easily dismissed, returning the user to their original document without disruption.
Objective
A user wants to quickly iterate on a button design in a Figma-like interface using PixelFlow.How the recipe adapts
The CursorHighlightAndSummon primitive would show the user selecting a button component. The ContextualMenuReveal would display options like 'Generate Variations', 'Optimize for Mobile', or 'Convert to React Component'. The AIExplanationModal would then show a loading state followed by a grid of new button designs or a generated code snippet, all within PixelFlow's clean UI, maintaining the original video's rapid, responsive feel.Scene durations (4 scenes, 15s)
Key Remotion components
Don't type, just speak. Flawless voice-to-text across every app at 240 WPM.

One workspace for knowledge scattered across Slack, email, Docs, and Notion. Connect your tools and build apps and agents that stay live with your data.

Notion positions itself as the backbone of your dock — a brand film of the Notion icon among everyday apps, ending on “Works with everything.”