Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计通过

markdown-to-storyboardMarkdown TO storyboard 笔记

Agent Skill

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

总安装

470

周安装

20

GitHub Stars

1,126

下载量

165
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:markdown-to-storyboard(Markdown TO storyboard 笔记)
来源仓库:https://github.com/moosegoose0701/skill-compose
仓库路径:skills/markdown-to-storyboard
安装命令:
npx skills add https://github.com/moosegoose0701/skill-compose --skill markdown-to-storyboard
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/moosegoose0701/skill-compose --skill markdown-to-storyboard

简介

该工具将 Markdown 文本转化为 CSV 格式的故事板。

  • 支持多种幻灯片布局与图像提示自动生成。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。
  • 适用于演示文稿设计与视频脚本可视化规划。
  • 输出包含演讲者注释与图像生成提示便于后续制作。
  • markdown-to-storyboard 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Markdown to Storyboard

Convert any markdown text into a structured storyboard CSV — the universal handoff format for downstream tools (slide builders, video editors, image generators).

Workflow

1. Analyze the Source Markdown

Read the markdown and identify:

  • Title and subtitle
  • Section headings (H2/H3) → natural slide boundaries
  • Key arguments / data points / quotes per section
  • Narrative arc: setup → development → climax → conclusion

2. Ask User Preferences

Prompt for (defaults in parentheses):

  • Slide count (auto: 8–15 based on content length, ~1 slide per 150–300 words)
  • Audience (general)
  • Tone (professional)
  • Language (same as source)

If user declines to specify, use defaults and proceed.

3. Assign Slide Types and Layouts

Available slide_type values:

TypePurposeTypical layout
coverOpening slidefull_bg
tocTable of contents (optional, 12+ slides)center_text
sectionChapter dividercenter_text
contentCore informationleft_img_right_text or top_img_bottom_text
quoteKey quote or statisticfull_bg
dataChart / number-driventwo_column
summaryClosing recapcenter_text or left_img_right_text
endThank-you / Q&A / CTAfull_bg

Available layout values: full_bg, left_img_right_text, top_img_bottom_text, center_text, two_column

4. Write the Image Prompt for Each Slide

For each slide, write an image_prompt (English, regardless of content language) that:

  • Describes a concrete scene, not an abstract concept
  • Includes a style keyword consistent across all slides (e.g. "flat vector illustration, soft pastel palette")
  • Specifies composition hints matching the layout (e.g. "wide shot, negative space on the right" for left_img_right_text)
  • Cover and end slides get more dramatic / visually impactful prompts

5. Draft Speaker Notes

speaker_notes column: 1–3 sentences of what the presenter would say. Leave empty if not requested.

6. Output the CSV

Write the storyboard to storyboard.csv using execute_code:

import csv

rows = [
    # [slide_no, slide_type, title, bullet_points, image_prompt, speaker_notes, layout]
]

header = ["slide_no", "slide_type", "title", "bullet_points", "image_prompt", "speaker_notes", "layout"]

with open("storyboard.csv", "w", newline="", encoding="utf-8") as f:
    writer = csv.writer(f)
    writer.writerow(header)
    writer.writerows(rows)

Column spec:

ColumnTypeDescription
slide_noint1-indexed
slide_typestrOne of the types above
titlestrSlide headline, ≤ 10 words
bullet_pointsstrNewline-separated (\n). Each ≤ 20 chars. Max 5 items
image_promptstrEnglish prompt for image generation
speaker_notesstrOptional presenter notes
layoutstrOne of the layouts above

7. Present for Review

Display the storyboard as a markdown table for the user to review. Wait for confirmation before the next pipeline step. If changes requested, update storyboard.csv and re-display.

Pacing Guidelines

  • Opening (slides 1–2): Hook — cover + bold opening statement or question
  • Body (slides 3 to N-2): One idea per slide. Alternate content and quote/data to vary rhythm
  • Closing (slides N-1 to N): Summary of key takeaways + end slide
  • Avoid consecutive slides of the same type. Insert a section divider between major parts.

Text Density Rules

  • Title: imperative or question form, ≤ 10 words
  • Bullets: ≤ 5 per slide, each ≤ 20 characters (CJK) / ≤ 12 words (Latin)
  • Prefer fragments over full sentences
  • Numbers stand alone: "3x faster" not "It is three times faster"

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.3%
按下载量换算55

Claude

32.35%
按下载量换算53

Cursor

18.84%
按下载量换算31

Gemini CLI

10.15%
按下载量换算17

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills