
toufuim/personal-ip-brand-intro-skill
ResearchThis page was translated automatically. Technical details may differ from the original. View the original
Open-source Codex Skill: Create personal IP brand openings with text, original illustrations, or user images, supporting uploaded music syncs and non-music custom beats.
Overview
An open-source Codex Skill designed to create personal IP brand opening animations with commercial value, clear pacing, and no fixed templates. Each production is redesigned according to the user's brand positioning by recreating the visual proposition, designing fingerprints, beat charts, storyboards, and animation compositions. - Official website: FuFu AI Lab - Threads: @xiaochou.chou - No logo or photo required; dynamic large text and original vector illustrations can be used for creation - Supports user-provided portraits, IP character images, product images, or logos - Supports mixing text, illustrations, and images - When users upload music, analyze rhythm, sections, and reshoot for photoshoot editing - If users have not uploaded music, design their own BPM and beat chart - Supports HyperFrames, Remotion, or explicitly specified dual-engine delivery - Supports 16:9, 9:16, and 1:1 - Storyboarding and inspection are completed first, and only rendered after user confirmation When users do not provide music, Skill selects BPM based on brand personality, text density, and video length, generating a unique audiomap.json beat chart. The finished product remains silent, but all text, illustrations, transitions, and branding finishes are still designed to the beat. Skill does not search, download, or secretly replace online music. If the user adds music later, the uploaded track will be reanalyzed and the beattable rebuilt, without directly using the original muted rhythm. 1. Organize brand name, character positioning, content areas, services, and slogans 2. Choose plain text, image-assisted, or mixed modes 3. Analyze user music or create an autonomous beat chart 4. Create brand-specific visual propositions and seven design fingerprints 5. Write rhythm storyboards accurate to the time point 6. Create using HyperFrames or Remotion 7. Check layout, dynamics, text, contrast, rhythm, and final video specifications 8. Render officially after obtaining user confirmation
Overview (machine translated)
An open-source Codex Skill designed to create personal IP brand opening animations with commercial value, clear pacing, and no fixed templates.
Each production is redesigned according to the user’s brand positioning by recreating the visual proposition, designing fingerprints, beat charts, storyboards, and animation compositions.
Author and Contact
- Official website: FuFu AI Lab
- Threads:@xiaochou.chou
Features
- No logo or photo is needed; you can create with dynamic large text and original vector illustrations
- Supports user-provided portraits, IP character images, product images, or logos
- Supports mixing text, illustrations, and images
- When users upload music, analyze rhythm, sections, and reshoot before performing a duet editing
- When users have not uploaded music, they can design their own BPM and beattable
- Supports HyperFrames, Remotion, or explicitly specified dual-engine delivery
- Supports 16:9, 9:16, and 1:1
- Complete the storyboarding and inspection first, then obtain user confirmation before officially rendering
How to handle it when there’s no music
When users do not provide music, Skill selects BPM based on brand personality, text density, and video length, generating a unique audiomap.json time chart.
The finished product remains silent, but all text, illustrations, transitions, and branding finishes are still designed to the beat. Skill does not search, download, or secretly replace online music.
If the user adds music later, the uploaded track will be reanalyzed and the beattable rebuilt, without directly using the original muted rhythm.
Installation Method
Method 1: Please install Codex
Paste the following content to Codex:
After installation, restart Codex.
Method 2: Manual installation
How to use
Plain text and illustrated version
Use people or IP images
Using Uploaded Music Sync
Three visual modes
| Mode | Applicable Scenarios |
|---|---|
text-illustration |
No external images, using dynamic text and original vector illustrations |
image-assisted |
Users can provide their person, IP, product image, or logo |
mixed |
Text, illustrations, and user images together form the brand system |
Main Production Process
- Organize brand name, role positioning, content areas, services, and slogans
- Choose plain text, image assistance, or blend mode
- Analyze user music or create an autonomous beat chart
- Establish a brand-exclusive visual proposition and seven design fingerprints
- Write rhythm storyboards that are precise down to the right moment
- Create using HyperFrames or Remotion
- Check layout, dynamics, text, contrast, rhythm, and final specifications
- After obtaining user confirmation, the render is officially done
System Requirements
- Node.js
- FFmpeg and FFprobe
- The default crafting path requires HyperFrames
- Remotion is only required when output is requested
Images and music uploaded by users are only kept in the user’s own production projects.
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add toufuim/personal-ip-brand-intro-skill