# Cursor Multitask / Subagents — /multitask in the new…
Let one session keep receiving tasks while ongoing work remains visible. Focus on order within parallel work rather than the number of windows.
```motion-demo
{
"id": "sequence",
"title": "Overall choreography",
"src": "demos/sequence.html",
"duration": 24.9,
"status": "authored",
"description": "Motion and composition example",
"initialTime": 3.8,
"params": [
{
"key": "title",
"label": "Sample main copy",
"type": "text",
"value": "Multitask",
"maxLength": 60
},
{
"key": "accent",
"label": "Accent color",
"type": "color",
"value": "#9966cc"
}
],
"presets": [
{
"label": "Default content",
"values": {
"title": "Multitask",
"accent": "#9966cc"
}
},
{
"label": "OpenDesign adaptation",
"values": {
"title": "OpenDesign",
"accent": "#7050da"
}
}
]
}
```
## Push into the entry point, then return to the whole view for execution
Let viewers read the tasks and parallel button first, then pull back to the complete app after clicking. Continue the main conversation downward while keeping the work-item panel fixed.
A white workspace, pale gray task blocks, and minimal accents support the flow. Use regular sans serif body text, slightly bolder task names, and compact status and time labels.
## Give each task its own identity and state
Keep the first task in the work list when adding the second. Let each independently queue, run, and complete; do not use one progress bar for all work.
When showing logs, expand only the focused item and retain brief states for others, avoiding four long logs competing for attention.
## Continued input demonstrates parallel capability
Enter a new request while tasks are still running, naturally showing that the session remains available. End in continued execution if appropriate; do not force every item into success.
For another business, prepare distinct content and state changes for each item, retaining a fixed list, continuous conversation, and independent feedback.
## Color and typography
Sans-serif interface text with a clear size hierarchy.
#f5f5f1 · #202122 · #9966cc · #f1ba8a · #9ab8ad
## Elements and production rules
App window / input field / parallel button / multiple tasks / status panel / scrolling conversation
- Add multiple tasks to the same session, with a fixed work-item panel showing the parallel task count
- push the window toward the entry point, pull back to the overview, then use scrolling text to show execution
- display running and completed tasks separately rather than depicting every in-progress item as successful
## Content replacement
- Adapt to any parallel-work tool; preserve one entry→multiple in-flight items→individual feedback
- Replace object text for the new business while keeping task-list identities stable
## 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": "V073",
"palette": [
"#f5f5f1",
"#202122",
"#9966cc",
"#f1ba8a",
"#9ab8ad"
],
"typography": "Sans-serif interface text with a clear size hierarchy",
"sample": "Multitask",
"objects": [
"App window",
"Input field",
"Parallel button",
"Multiple tasks",
"Status panel",
"Scrolling conversation"
],
"frames": [
{
"path": "assets/reference-1.jpg",
"timeSeconds": 0.0,
"frameId": "K0001",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-2.jpg",
"timeSeconds": 3.0,
"frameId": "K0013",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-3.jpg",
"timeSeconds": 5.0,
"frameId": "K0020",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-4.jpg",
"timeSeconds": 11.0,
"frameId": "K0040",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-5.jpg",
"timeSeconds": 14.0,
"frameId": "K0048",
"role": "reference-evidence",
"status": "measured"
},
{
"path": "assets/reference-6.jpg",
"timeSeconds": 22.0,
"frameId": "K0068",
"role": "reference-evidence",
"status": "measured"
}
],
"stages": [
{
"start": 0.0,
"label": "Task input",
"end": 1.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 1.0,
"label": "Push into the parallel button",
"end": 3.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 3.0,
"label": "Start parallel work",
"end": 5.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 5.0,
"label": "Agent response",
"end": 8.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 8.0,
"label": "Second task",
"end": 11.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 11.0,
"label": "Four work items",
"end": 14.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 14.0,
"label": "Task log",
"end": 18.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 18.0,
"label": "Follow-up actions",
"end": 22.0,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
},
{
"start": 22.0,
"label": "Continue execution",
"end": 24.9,
"status": "inferred",
"basis": "Representative-frame review; approximate boundaries"
}
],
"duration": 24.9,
"fontFamily": "Arial, sans-serif"
}
```