← 返回 Skills 市场
zrxparley

PPT Layout Matcher

作者 Aha.Gare · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
34
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install ppt-layout-matcher
功能描述
Intelligent PPT layout matching system based on a 45-page design layout library (16:9). Analyzes input content and automatically recommends the best slide la...
使用说明 (SKILL.md)

PPT Layout Matcher

Intelligent layout matching for PowerPoint presentations based on a 45-page professional design layout library.

When to Use

  • User wants to create a PPT from content
  • User asks "which layout should I use for this content?"
  • User provides bullet points / data / flow descriptions and wants them turned into slides
  • User mentions "PPT", "presentation", "slides", "版式", "排版"

Workflow

Step 1: Analyze Content Structure

Extract key features from user input:

  • Number of items (3 key metrics? 4 steps? 2 comparisons?)
  • Content nature (data/metrics, process/steps, comparison, gallery, list)
  • Presence of numbers/charts/images
  • Timeline/step indicators

Step 2: Run Matching Algorithm

Execute the bundled script to get layout recommendations:

python3 ppt_layout_matcher.py

Or use the Python API:

from ppt_layout_matcher import recommend_layout
results = recommend_layout("user content here", top_k=3)
for template, score, analysis in results:
    print(f"{template.name} (Slide {template.slide_ref}) - Score: {score}")

Step 3: Present Top 3 to User

Show:

  • Layout name and reference slide number
  • Match score
  • Structure description
  • Why this layout was recommended

Let the user confirm before creating.

Step 4: Create PPT with python-pptx

Once confirmed, create the slide:

  • 16:9 dimensions (13.33" × 7.50")
  • Dark theme (deep blue/gray background)
  • White text, accent colors for data highlights
  • Follow the reference layout structure from references/layout_templates.md

Step 5: QA

Verify with markitdown:

python -m markitdown output.pptx | head -100

Layout Categories (Quick Reference)

Category Count Use Case
A-Cover 2 Report cover, proposal front page
B-Divider 2 Section transitions
C-Data Display 9 Metrics, charts, data highlights
D-Process/Steps 6 STEP flows, development stages
E-Comparison 5 A/B comparison, multi-solution, SWOT
F-Grid/Cards 8 2×2, 3×2, 4×N card grids
G-Gallery 1 Image gallery, timeline
H-List/Points 4 Numbered lists, text+image lists
I-Infographic 1 One Pager overview
J-Special 7 Radial, wraparound, mixed layouts

Design Conventions

  • Title bar: top 0.3-1.0" area
  • Content area: 1.2"-7.0"
  • Footer: 6.5"-7.2"
  • Margins: ≥0.5" on all sides
  • Heading: 20-30pt Bold, Body: 10-14pt, Big Numbers: 36-60pt Bold
  • Dark background, white text, accent colors for highlights

References

Full layout library documentation: references/layout_templates.md

安全使用建议
This skill looks safe for its stated purpose. Before installing, be aware that it may run local Python code and may require installing python-pptx; use trusted package sources and review the script if your environment restricts local code execution.
功能分析
Type: OpenClaw Skill Name: ppt-layout-matcher Version: 1.0.0 The PPT Layout Matcher skill bundle is a legitimate tool designed to analyze text content and recommend PowerPoint layouts using a predefined library of 45 templates. The core logic in `ppt_layout_matcher.py` uses regular expressions to identify content features (e.g., numbers, processes, comparisons) and calculate a matching score for various slide structures. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found; the code is well-documented and its behavior is strictly aligned with the stated purpose of assisting in presentation design.
能力评估
Purpose & Capability
The artifacts consistently describe a PPT layout matching assistant that analyzes content, recommends layouts, and optionally helps create slides using the documented layout library.
Instruction Scope
The skill tells the agent to run bundled Python code and to present recommendations before slide creation; this is disclosed and purpose-aligned, but users should recognize that local code may be executed.
Install Mechanism
There is no install spec or declared required binary/package, while the README documents installing python-pptx and SKILL.md references markitdown for QA. This is not suspicious, but dependency installation is under-declared in metadata.
Credentials
The shown artifacts do not request credentials, network access, privileged paths, broad filesystem indexing, or account mutation. Local PPT output creation is proportionate to the stated purpose.
Persistence & Privilege
No background workers, persistence mechanisms, auth/session/profile access, or elevated privileges are shown.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install ppt-layout-matcher
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /ppt-layout-matcher 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of PPT Layout Matcher. - Analyzes user content and recommends optimal slide layouts from a 45-page (16:9) professional design library, organized into 10 categories. - Matches content structure (e.g., steps, metrics, comparisons) to layout types. - Provides top 3 recommended layouts with scores and explanations. - Generates slides using python-pptx with consistent dark theme and design conventions. - Includes quick reference for layout categories and workflow for review and QA.
元数据
Slug ppt-layout-matcher
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

PPT Layout Matcher 是什么?

Intelligent PPT layout matching system based on a 45-page design layout library (16:9). Analyzes input content and automatically recommends the best slide la... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 34 次。

如何安装 PPT Layout Matcher?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ppt-layout-matcher」即可一键安装,无需额外配置。

PPT Layout Matcher 是免费的吗?

是的,PPT Layout Matcher 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

PPT Layout Matcher 支持哪些平台?

PPT Layout Matcher 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 PPT Layout Matcher?

由 Aha.Gare(@zrxparley)开发并维护,当前版本 v1.0.0。

💬 留言讨论