ııı motion.mdVideo style specification
Higgsfield / V305

Higgsfield — plugin now in the ChatGPT app

Move from the plugin entry point to usable batch outputs, then use performance data to determine the next action. Make generation and selection one continuous workflow.

Composition and choreography Playable · Replaceable content
Overall choreographyRule example
0.00 / 35.30 seconds
Motion and composition example Open separately ↗
View HTML source
<!doctype html><html lang="zh-CN"><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"><title>Higgsfield — plugin now in the ChatGPT app · sequence</title><style>html,body{margin:0;width:100%;height:100%;overflow:hidden;background:#ffffff}canvas{display:block;width:100%;height:100%;object-fit:contain}</style><body data-mode="sequence"><canvas width="1000" height="600"></canvas><script src="data.js"></script><script src="common.js"></script><script src="scene.js"></script><script src="bridge.js"></script></body></html>
Visual elementsPlugin selectionModel listTask inputBatch resultsChartRanking card
MOTION LANGUAGE

Motion and composition

Local motion · Composition rules

01

Establish where the plugin can be used first

Set the context with platform badges, plugin entry points, and model or skill lists. Use clear sans-serif names and an accent background for selection rather than low-contrast colored text; reduce emphasis on other options.

White, black text, and fluorescent green define the main hierarchy; use pink and pale blue only to distinguish media or result categories.

02

One batch request, one complete set of outputs

Specify product source, creative direction, and output count in the input. Then show task progress before opening an image or video grid; its contents should vary while clearly belonging to the same batch.

03

From gallery to performance ranking

From gallery to performance rankingRule example
0.00 / 14.00 seconds
20–34 seconds · Motion example Open separately ↗
View HTML source
<!doctype html><html lang="zh-CN"><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"><title>Higgsfield — plugin now in the ChatGPT app · sequence</title><style>html,body{margin:0;width:100%;height:100%;overflow:hidden;background:#ffffff}canvas{display:block;width:100%;height:100%;object-fit:contain}</style><body data-mode="clip" data-start="20" data-end="34"><canvas width="1000" height="600"></canvas><script src="data.js"></script><script src="common.js"></script><script src="scene.js"></script><script src="bridge.js"></script></body></html>

Show finished work before metrics and rankings. When highlighting a leaderboard item, retain its identity or thumbnail rather than showing bars disconnected from the work.

The final automation should follow from the selection, such as iterating on a high-performing direction or scheduling another production round.

04

Change the task and its evaluation criteria

Adapt to product images, campaign videos, or design proposals. Rewrite prompts, asset catalog, output set, metrics, and next actions together; different tasks need not share the same business metric.

SEQUENCE

Sections and rhythm

Click a section to seek to it in the choreography example above.

Composition reference at 0.00 seconds
0.00s
Composition reference at 5.00 seconds
5.00s
Composition reference at 12.00 seconds
12.00s
Composition reference at 20.00 seconds
20.00s
Composition reference at 26.00 seconds
26.00s
Composition reference at 32.00 seconds
32.00s

Visual references · Discrete still frames

KEEP THE RELATIONSHIPS

Creating other themes

  • Can be adapted to a design plugin's generation→publishing→analysis workflow
  • Keep rankings and metrics consistent when replacing data and work

Production constraints

Plugin selection / Model list / Task input / Batch results / Charts / Ranking cards

  • Select the plugin and skill before entering a cross-tool task, then reveal execution steps in order
  • results change from an asset grid to performance metrics and rankings, showing an operational loop after generation
  • finally, turn the one-off task into a recurring workflow