Communitygithub.com

PAISHU-AI/PAI-zong-NB-v5.8

Create HTML-based video compositions, animated title cards, social overlays, captioned talking-head videos, audio-reactive visuals, and shader transitions using HyperFrames. HTML is the source of truth for video. Use when the user wants a rendered MP4/WebM from an HTML composition, wants to animate text/logos/charts over media, needs captions synced to audio, wants TTS narration, or wants to convert a website into a video.

What is PAI-zong-NB-v5.8?

PAI-zong-NB-v5.8 is a Codex agent skill that create HTML-based video compositions, animated title cards, social overlays, captioned talking-head videos, audio-reactive visuals, and shader transitions using HyperFrames. HTML is the source of truth for video. Use when the user wants a rendered MP4/WebM from an HTML composition, wants to animate text/logos/charts over media, needs captions synced to audio, wants TTS narration, or wants to convert a website into a video.

Works with~Claude Code✓Codex CLI~Cursor
npx skills add https://github.com/PAISHU-AI/PAI-zong-NB-v5.8/tree/HEAD/skills/hyperframes

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

hyperframes

Purpose

Create HTML-based video compositions, animated title cards, social overlays, captioned talking-head videos, audio-reactive visuals, and shader transitions using HyperFrames. HTML is the source of truth for video. Use when the user wants a rendered MP4/WebM from an HTML composition, wants to animate text/logos/charts over media, needs captions synced to audio, wants TTS narration, or wants to convert a website into a video.

Use When

  • The user request matches hyperframes, creative or the skill description.
  • Create HTML-based video compositions, animated title cards, social overlays, captioned talking-head videos, audio-reactive visuals, and shader transitions using HyperFrames. HTML is the source of truth for video. Use when the user wants a rendered MP4/WebM from an HTML composition, wants to animate text/logos/charts over media, needs captions synced to audio, wants TTS narration, or wants to convert a website into a video.
  • The task benefits from domain-specific workflow, checks, output structure, or safe fallback behavior.

Workflow

  1. Clarify target artifact, audience, style, platform, dimensions, and source material.
  2. Create a production-ready brief, prompt, storyboard, or asset plan before tool execution.
  3. Use image/video/design tools only when available in the current session.
  4. If tools are unavailable, return portable prompts/specs and state NOT RUN.

Codex Runtime Rules

  • Use only tools, plugins, MCP servers, files, commands, and credentials that are actually available in the current Codex session.
  • If an external write, install, publish, deploy, paid action, credential use, account operation, physical-device action, or production action is needed, ask for explicit confirmation first.
  • If the required tool is unavailable, return a portable fallback such as Markdown, CSV, prompt pack, implementation plan, config sketch, or verification checklist.
  • Do not claim files, media, issues, API calls, payments, deployments, transactions, messages, or integrations were created unless there is concrete local or tool evidence.
  • Treat source repository runtime details as historical source context, not active Codex capabilities.

Avoid

  • Do not claim media was generated unless a real artifact exists.
  • Do not bypass copyright, likeness, or safety constraints.

Output

Return the useful artifact first, then evidence and verification status.

For non-trivial work, include:

Result: ...
Evidence: ...
Verification: PASS/FAIL/NOT RUN
Risks / Next Steps: ...

Related Skills