ppt-reader

Use when user sends a PPT/PPTX file or asks to read PowerPoint content. 适用于:读取PPT文件、解析演示文稿内容、提取幻灯片文本。

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "ppt-reader" with this command: npx skills add jingqiu2180/ppt-reader

PPT Reader

Overview

读取和解析 PowerPoint 文件(.pptx),提取幻灯片文本内容。

When to Use

  • 用户发送了 .pptx 文件
  • 需要提取PPT中的文字内容
  • 需要了解PPT的结构和页数

Quick Reference

操作命令
查看PPT页数unzip -l file.pptx | grep "slide[0-9]*\.xml" | wc -l
提取所有文本unzip -p file.pptx "ppt/slides/slide*.xml" | sed 's/<[^>]*>//g'
逐页提取见下方脚本

Implementation

方法1:快速提取所有文本

unzip -p "file.pptx" "ppt/slides/slide*.xml" 2>/dev/null | sed 's/<[^>]*>//g' | tr -s ' \n'

方法2:逐页提取(推荐)

cd /path/to/ppt/
for i in {1..N}; do 
  echo "=== Slide $i ===" 
  unzip -p "file.pptx" "ppt/slides/slide$i.xml" 2>/dev/null | sed 's/<[^>]*>//g' | tr -s ' \n'
  echo ""
done

方法3:获取PPT基本信息

# 页数
unzip -l "file.pptx" | grep -c "slide[0-9]*\.xml"

# 文件结构
unzip -l "file.pptx" | grep -E "slide[0-9]+\.xml"

Workflow

  1. 确认文件路径 - 从 /root/.openclaw/media/inbound/ 获取文件
  2. 获取页数 - 确定有多少张幻灯片
  3. 逐页提取 - 循环提取每页内容
  4. 整理输出 - 汇总成结构化摘要

Common Patterns

处理飞书发送的PPT

飞书发送的文件通常保存在:

/root/.openclaw/media/inbound/

文件名格式:

原始文件名-uuid.pptx

提取特定幻灯片

# 只提取第1页
unzip -p "file.pptx" "ppt/slides/slide1.xml" | sed 's/<[^>]*>//g'

# 提取第1-5页
for i in {1..5}; do unzip -p "file.pptx" "ppt/slides/slide$i.xml" | sed 's/<[^>]*>//g'; done

Limitations

  • 只能提取文本内容,无法提取图片、图表
  • 无法获取格式信息(字体、颜色、布局)
  • 复杂表格可能提取不完整
  • 仅支持 .pptx 格式(不支持旧版 .ppt)

Real-World Example

# 完整提取流程
PPT_FILE="/root/.openclaw/media/inbound/智能体-概述-20260116-xxx.pptx"

# 1. 获取页数
PAGE_COUNT=$(unzip -l "$PPT_FILE" | grep -c "slide[0-9]*\.xml")
echo "Total slides: $PAGE_COUNT"

# 2. 逐页提取
for i in $(seq 1 $PAGE_COUNT); do
  echo "=== Slide $i ==="
  unzip -p "$PPT_FILE" "ppt/slides/slide$i.xml" 2>/dev/null | sed 's/<[^>]*>//g' | tr -s ' \n'
  echo ""
done

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

Img2img

Generate images from text descriptions using DALL-E 3 while adhering to usage policies and avoiding realistic human faces.

Registry SourceRecently Updated
General

Habitat-GS-Navigator

Navigate and interact with photo-realistic 3DGS environments via the Habitat-GS Bridge. Use when: user asks to explore a 3D scene, perform embodied navigatio...

Registry SourceRecently Updated
General

Memory Palace

持久化记忆管理。Use when: 用户告诉你个人信息/偏好/习惯、需要记住项目状态/技术决策、完成任务后有可复用经验、用户说"记住""别忘了""下次注意"、需要回忆之前的对话内容。支持语义搜索和时间推理。

Registry SourceRecently Updated
General

Podcast Transcript Mining Authority Positioning

Extract guest appearances, speaking topics, and soundbites from podcast transcripts to build authority portfolios and generate podcast pitch templates. Use w...

Registry SourceRecently Updated