菜单
Generate concise hand-drawn whiteboard infographic prompts, layout notes, and accessibility text for AI image generation. Use this as the creator-facing prompt bridge inside the AI whiteboard infographic video pipeline.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/huige-opc/whiteboard-speak --skill hand-drawn-infographic-creator命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Generate editable HyperFrames whiteboard infographic video projects from board_manifest.json, motion_plan.json, voiceover_segments.json, and board visual assets such as board.png or asset.kind=url. Use when Codex needs to synthesize edge-tts Chinese narration, measure real segment durations, update motion timing, create captions, render MP4 previews, and extract action-level start/done keyframes for AI whiteboard explanation videos.
Create single-board or multi-board video-ready control packages for hand-drawn whiteboard infographic explainer videos. Use when Codex needs to turn board_spec.json plus board.png, or a project package with board_asset_manifest.json, infographic_plan.json, board_specs, and voiceover_segments.json, into precise board_manifest.json, motion_plan.json, annotation_manifest.json, board_index.json, SVG/HTML alignment layers, and annotation keyframe checks for cursor underline, circle, box, check, and strike effects.
Orchestrate the AI whiteboard infographic explainer video pipeline from a natural-language user request or a topic/script input file into a validated project package. Use when the user says things like "请使用白板总编排skill帮我做一个视频", "白板总编排 Skill", "用白板信息图流水线做视频", gives a topic such as "主题为..." and a duration such as "30-60秒左右", or asks Codex to make an AI 白板信息图讲解视频. Coordinate ip-cognition-script-polisher, ip-hand-drawn-infographic-planner, hand-drawn-infographic-creator, automatic or interactive model-image handoff, hand-drawn-infographic-video-board, and whiteboard-infographic-video-renderer to produce script assets, infographic plans, creator prompts, board_asset_manifest.json, board control packages, AI narration, captions, HyperFrames, preview.mp4, keyframes, and integration_report.md without substituting placeholder images.
基于 SOC 职业分类
正在显示 SKILL.md
| name | hand-drawn-infographic-creator |
| description | Generate concise hand-drawn whiteboard infographic prompts, layout notes, and accessibility text for AI image generation. Use this as the creator-facing prompt bridge inside the AI whiteboard infographic video pipeline. |
Turn a semantic board prompt into a final image-generation prompt for a sparse hand-drawn whiteboard infographic. This skill produces prompt text and review notes only. It does not create video, audio, animation coordinates, or final board-control JSON.
For each board, write a creator output file and a final image prompt:
creator_outputs/board-XX.creator_output.md
imagegen_prompts/board-XX.imagegen_prompt.txt
The final prompt is the input for an image model. If the image tool returns only a preview, the operator must manually save the PNG into:
images/board-XX.model-generated.png
Palette:
background: "#faf8f3"
primary_line: "#1a2332"
annotation: "#2d5a7b"
positive: "#4a9d9e"
risk: "#e63946"
insight: "#f4a261"
Use one or two semantic highlight colors per board. Use color for meaning, not decoration.
Every final prompt should include:
16:9 landscape.内容简洁一点，不要逐字写满口播.photorealistic, 3D render, CGI, stock photo, corporate dashboard, smooth digital art, dense tiny text, watermark.workflow_asset, keep that id in creator notes.board_asset_manifest.json.# Creator Output: board-01
## Board Intent
[One sentence explaining the board.]
## Required Visible Labels
- [short label]
- [short label]
## Layout Notes
- [where the main object should sit]
- [where the margin note or takeaway should sit]
## Final Image Prompt
[model-ready prompt]
## Negative Prompt
[negative prompt]
## Review Notes
- Text that must be checked after generation.
- Elements that need D calibration if the model drifts.