ppt制作-高效版

SKILL效率
效率 未标注权限 许可证不明确
安装这个技能

第一核心功能是“HTML替代PPT(高效快)”,第二核心功能是“风格提取与UI提示词生成(提升视觉表现)”。其第一核心功能通过HTML直接替代传统PPT制作流程,以模块化网页结构快速生成用于汇报与讲解的演示内容,使整体制作过程显著更快、更轻量,并将制作效率提升约80%;第二核心功能基于用户输入的参考图片或风格信息,自动提取视觉语言并生成可复用的前端UI设计提示词系统,用于统一控制后续HTML页面的配色、布局与组件规范,从而在保持高效率的同时提升整体视觉一致性与设计表达能力。

一句话选型结论

按资料字段整理

ppt制作-高效版是效率类的 Agent Skill,安装方式未标注,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类效率 安装方式未标注 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

未标注
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install user_909957f0/tutorial-website-builder
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

效率
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方user_909957f0仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式未标注安装包 community
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791710503
正文体量7.5 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
整理待办与日程

把零散任务收进清单,排出优先级。

批量处理重复任务

把固定动作做成模板,重复执行。

把流程写成清单

把口头流程整理成可复用的步骤。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 没有做站内核验,内容以作者仓库为准。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在效率场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
ppt制作-高效版当前 未标注权限 未标注 未标注 未标注 有
Nano PDF 文档 联网访问 ClawHub 市场 424.9 万 OpenClaw 有
自动更新程序 浏览器控制 ClawHub 市场 296.4 万 OpenClaw 有
Gemini 助手 联网访问 ClawHub 市场 254.7 万 OpenClaw 有
word DOCX 文档 联网访问 ClawHub 市场 239.6 万 OpenClaw 有
Excel XLSX 前端 联网访问 ClawHub 市场 227.0 万 OpenClaw 有

SKILL.md 全文

7.5 KB
模型实际读到的说明书原文,来自技能仓库

Tutorial Website Builder

Overview

Use this skill for two related workflows:

  1. Extract a reusable UI style prompt from a user-provided reference image.
  2. Generate a single-file paged HTML tutorial using a selected UI style prompt.

The expected HTML output is an immediately usable tutorial page, not a marketing landing page. The first screen must enter the tutorial content directly.

Workflow A: Extract Style From Image

When the user provides a reference image and asks to create, extract, save, or package a style:

  1. Read references/style-extraction-schema.md.
  2. Analyze only the visual language of the image: layout, typography, color, components, spacing, media treatment, illustration logic, interaction mood, and responsive implications.
  3. Fill the schema fields with concrete reusable defaults. Do not merely describe the image; convert what you see into a style system another agent can apply to a webpage.
  4. Name the style by its visual language, using lowercase hyphen-case English for the file name.
  5. Save the reusable prompt as references/styles/<style-name>.md.
  6. If the image contains protected logos, characters, private UI, or copyrighted artwork, extract the general visual system only. Do not instruct future pages to copy exact proprietary elements.
  7. Report the created style file path and the human-readable style name.

Do not generate the tutorial HTML in this workflow unless the user also asks for it.

Workflow B: Generate Tutorial HTML

When the user asks to generate or modify a tutorial HTML:

  1. Identify available style prompts:
    - references/default-ui-system.md
    - every *.md file in references/styles/
  2. If the user has not specified a style, ask which style prompt to use. Keep the question concise and include the default option.
  3. Read the selected style prompt before designing the HTML.
  4. Also read references/production-guide.md.
  5. Always apply the three bottom-level logic rules below, even if the selected style prompt does not mention them.
  6. Generate or edit the HTML according to the selected style, tutorial content, and media available in the project.

Default media folders remain:

tutorial.html
images/
  image_01.png
  image_02.png
videos/
  video_01.mp4

The user may confirm or adjust actual image/video placement later in the conversation. Do not require that confirmation before producing a scaffold when the user asks for one.

Required Bottom-Level Logic

These rules override individual style prompts and component convenience.

1. No Card-Inside-Card Composition

Never build the page by stacking many cards inside other cards. Use one primary page/container surface, then create hierarchy with grid, typography, borders, divider lines, labels, media frames, verification blocks, and whitespace.

Avoid repeated nested rounded rectangles such as card > card > panel > media card.

2. Meaningful Graphics Only

Do not add graphics by default. Any illustration, icon, shape, branch, leaf, animal, smoke, stream, hill, background mark, or decorative object must have a clear purpose:

  • Aesthetic framing.
  • Topic symbolism.
  • Section rhythm.
  • Directional guidance.
  • State feedback.
  • Media annotation.
  • Brand mood reinforcement.

If the graphic cannot be explained, omit it. If graphics make the page busy or weaken the focal point, remove them.

3. Tutorial Readability First

Layout styles can be diverse, but the page is a tutorial first. Every page must make these items easy to find:

  • Current step.
  • Current action.
  • Why the step matters.
  • Media that verifies the action.
  • Observable success condition.

Do not use visual variation that makes reading order ambiguous, separates media from its explanation, hides verification standards, creates long scanning paths, or turns the tutorial into a decorative poster.

User Configuration

Before generating the final HTML, ask whether any selected style defaults should be changed. Keep the question concise:

默认使用所选风格提示词。是否需要修改主色、字体、圆角、间距、导航、媒体样式、插图风格或动效?如果不需要,我将按默认值生成。

If the user does not specify changes, proceed with the selected style defaults. Do not repeatedly ask configuration questions.

When the user provides overrides, update only the relevant tokens or component rules while preserving the rest of the selected style.

Tutorial Content Rules

  1. Inventory the user-provided material:
    - Steps or rough workflow text.
    - Screenshots in images/, preferably named image_01.png, image_02.png, etc.
    - Videos in videos/, preferably named video_01.mp4, video_02.mp4, etc.
    - Official sites, source articles, tool pages, documentation links, or other external references.
  2. Split the tutorial by the real execution order. Each page must cover one action or one tightly related group of actions.
  3. For each page, write only the current step:
    - What the user is trying to accomplish.
    - What to do now.
    - What visible result means the step is complete.
  4. Attach media after the explanation so the image or video verifies the operation, rather than replacing the instructions.
  5. Number all figures globally by appearance order: 图 1, 图 2, 图 3. Do not reset numbering per step.
  6. Embed links directly in the body text with target="_blank" and rel="noopener noreferrer".

Output Requirements

  • Create a single HTML file unless the user explicitly asks for a framework project.
  • Use assets/newspaper-tutorial-template.html as a structural starting point when useful, but restyle it according to the selected style prompt.
  • Keep all style choices in CSS custom properties or equivalent design tokens.
  • Keep exactly one navigation area near the top: current page indicator plus previous/next buttons.
  • Use data-step sections and a single totalSteps value in JavaScript.
  • Derive Chinese page numerals from the step count.
  • Support clickable image enlargement. Videos should remain playable inline.
  • Make mobile layout single-column with large enough navigation buttons.

Quality Gate

Before delivery, check:

  • The selected style prompt is applied consistently.
  • The three bottom-level logic rules are applied.
  • Previous/next buttons correctly disable on first and last pages.
  • Only one navigation area exists.
  • Current page indicator and total page count match the number of step sections.
  • Figure numbers increase globally in display order.
  • Every src path points to the correct images/ or videos/ directory.
  • Every image has useful alt text and every media item has a concise caption.
  • Links open in a new tab and use rel="noopener noreferrer".
  • The page works when opened as a local HTML file.
  • Mobile width does not create horizontal scrolling, clipped buttons, or text overlap.

常见问题

5 问
答案由本站依据条目字段整理
ppt制作-高效版怎么安装?

在支持 Skills 的客户端里执行:skillhub install user_909957f0/tutorial-website-builder。安装方式为未标注。装完刷新技能列表或重开客户端。

ppt制作-高效版支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

ppt制作-高效版权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

ppt制作-高效版收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

ppt制作-高效版用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。