# Cursor Cloud Agents
Introduce product updates through brief claims followed by brief actions. Give each feature its own objects; unify the reading pattern, not interface content.
```motion-demo
{
"id": "sequence",
"title": "Overall choreography",
"src": "demos/sequence.html",
"duration": 43.866667,
"status": "authored",
"description": "Motion and composition example",
"initialTime": 8.2,
"params": [
{
"key": "title",
"label": "Sample main copy",
"type": "text",
"value": "Agent UX improvements",
"maxLength": 60
},
{
"key": "accent",
"label": "Accent color",
"type": "color",
"value": "#9966cc"
}
],
"presets": [
{
"label": "Default content",
"values": {
"title": "Agent UX improvements",
"accent": "#9966cc"
}
},
{
"label": "OpenDesign adaptation",
"values": {
"title": "OpenDesign",
"accent": "#7050da"
}
}
]
}
```
## One sentence on black, one example on white
Center the feature explanation on black in simple sans serif. Finish and hold the sentence for reading before cutting to the pale interface; contrast naturally creates chapter boundaries.
Reserve accents for mode labels, current commands, and status, avoiding a new decorative system for each feature.
## Four features, four concrete actions
Show requests and confirmations for thread subscriptions, labels and later input for custom modes, independent work items for subagents, and goal commands with ongoing tasks for cross-turn goals. Do not turn every chapter into the same input box with different text.
| Update direction | Corresponding interaction view |
| --- | --- |
| Task subscription | Thread request and subsequent response |
| Custom mode | Mode label and input |
| Isolated subagents | Multiple independent task states |
| Persistent goal | Goal command and subsequent work |
## Summarize only what has already been explained
Collect feature names into a final checklist, followed by the brand and update entry; do not add an undemonstrated fifth feature.
Rewrite other version updates with the same two-beat structure. Sentences may vary in length, but leave enough time to understand one action in each interaction segment.
## Color and typography
Sans-serif interface text with a clear size hierarchy.
#f5f5f1 · #202122 · #9966cc · #f1ba8a · #9ab8ad
## Elements and production rules
Explanation on black / interaction on white / thread reply / mode label / subagent list / goal command / summary checklist
- Present each feature as a one-sentence message on black followed by a short interaction on white, maintaining a regular rhythm
- different features require subscription, mode, parallelism and target objects respectively rather than shared fake data
- the closing list repeats demonstrated items without introducing new capabilities
## Content replacement
- For changelogs, retain proposition→interaction evidence as two beats for each item
- Replace the feature set, ensuring each claim matches the actual action that follows
## Production parameters
[Visuals and objects](style.json) · [Section timing](timeline.json) · [Position data](tracks.json) · [File manifest](manifest.json)
This specification is based on the reference video. The HTML provides editable choreography examples; each file identifies the source, precision and valid interval of its data.
```motion-board
{
"version": 1,
"brand": "Cursor",
"id": "V260",
"palette": [
"#f5f5f1",
"#202122",
"#9966cc",
"#f1ba8a",
"#9ab8ad"
],
"typography": "Sans-serif interface text with a clear size hierarchy",
"sample": "Agent UX improvements",
"objects": [
"Explanation on black",
"Interaction on white",
"Thread reply",
"Mode label",
"Subagent list",
"Goal command",
"Summary checklist"
],
"frames": [
{
"path": "assets/reference-1.jpg",
"timeSeconds": 0.0,
"frameId": "K0001",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-2.jpg",
"timeSeconds": 7.0,
"frameId": "K0017",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-3.jpg",
"timeSeconds": 15.0,
"frameId": "K0034",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-4.jpg",
"timeSeconds": 27.0,
"frameId": "K0058",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-5.jpg",
"timeSeconds": 34.0,
"frameId": "K0072",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-6.jpg",
"timeSeconds": 41.0,
"frameId": "K0088",
"role": "reference-evidence",
"status": "measured"
}
],
"stages": [
{
"start": 0.0,
"label": "Update heading",
"end": 3.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 3.0,
"label": "Task-subscription explanation",
"end": 7.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 7.0,
"label": "Thread subscription",
"end": 12.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 12.0,
"label": "Custom-mode explanation",
"end": 15.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 15.0,
"label": "Mode input",
"end": 19.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 19.0,
"label": "Isolated-subagent explanation",
"end": 24.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 24.0,
"label": "Parallel tasks",
"end": 27.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 27.0,
"label": "Cross-turn goal explanation",
"end": 31.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 31.0,
"label": "Goal command",
"end": 34.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 34.0,
"label": "Noninterrupting guidance explanation",
"end": 38.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 38.0,
"label": "Feature summary",
"end": 41.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 41.0,
"label": "Brand",
"end": 43.866667,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
}
],
"duration": 43.866667,
"fontFamily": "Arial, sans-serif"
}
```