with one click
summarize
快速命令行工具,用于生成URL、本地文件和YouTube链接的摘要。支持网页、PDF、图像、音频等多种格式。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
快速命令行工具,用于生成URL、本地文件和YouTube链接的摘要。支持网页、PDF、图像、音频等多种格式。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Word文档处理:创建、编辑、读取.docx文件。用户提到"Word"、"文档"、".docx"、"报告"、"信函"时使用。
使用腾讯云 API 密钥调用腾讯云各种服务,包括翻译、CVM、COS等。
图片识别与图文理解,基于 PaddleOCR-VL 模型,支持识别图片中的文字、场景、物体等信息。支持109种语言,可输出JSON或Markdown格式。
You MUST use this for gathering contexts before any work. This is a Knowledge management for AI agents. Use `brv` to store and retrieve project patterns, decisions, and architectural rules in .brv/context-tree. Uses a configured LLM provider (default: ByteRover, no API key needed) for query and curate operations.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Now with WAL Protocol, Working Buffer, Autonomous Crons, and battle-tested patterns. Part of the Hal Stack 🦞
| name | summarize |
| description | 快速命令行工具,用于生成URL、本地文件和YouTube链接的摘要。支持网页、PDF、图像、音频等多种格式。 |
| triggers | ["摘要","summarize","总结","内容摘要","概括","内容总结","文章摘要","视频摘要"] |
| homepage | https://summarize.sh |
| metadata | {"clawdbot":{"emoji":"🧾","requires":{"bins":["summarize"]},"install":[{"id":"brew","kind":"brew","formula":"steipete/tap/summarize","bins":["summarize"],"label":"Install summarize (brew)"}]}} |
快速命令行工具,用于生成URL、本地文件和YouTube链接的摘要。支持网页、PDF、图像、音频等多种格式。
# 摘要网页
summarize https://example.com/article
# 摘要本地文件
summarize document.pdf
summarize report.docx
summarize presentation.pptx
# 摘要YouTube视频
summarize https://www.youtube.com/watch?v=example
# 摘要图像文件
summarize image.jpg
# 指定输出长度
summarize https://example.com --length short # 简短摘要
summarize https://example.com --length medium # 中等长度
summarize https://example.com --length long # 详细摘要
summarize https://example.com --length xl # 非常详细
summarize https://example.com --length xxl # 最详细
# 指定字符数
summarize https://example.com --length 500 # 约500字符
summarize https://example.com --length 1000 # 约1000字符
# 基本网页摘要
summarize https://news.example.com/article123
# 使用高级提取(针对复杂网站)
summarize https://example.com --firecrawl always
# 仅提取内容不生成摘要
summarize https://example.com --extract-only
# 结构化输出
summarize https://example.com --json
# PDF文件
summarize document.pdf
summarize report.pdf --length long
# Word文档
summarize report.docx
summarize proposal.docx --json
# PowerPoint演示文稿
summarize presentation.pptx
summarize slides.pptx --length medium
# 图像文件(OCR提取)
summarize screenshot.png
summarize diagram.jpg --length short
# 音频文件(语音转文字后摘要)
summarize podcast.mp3
summarize interview.m4a --length medium
# 基本视频摘要
summarize https://www.youtube.com/watch?v=dQw4w9WgXcQ
# 使用Apify服务(需要API密钥)
export APIFY_API_TOKEN=your_token
summarize https://youtube.com/video --youtube auto
# 指定视频特定部分
summarize https://youtube.com/video#t=60s,300s # 摘要1-5分钟部分
创建 ~/.summarize/config.json:
{
"model": "openai/gpt-5.2",
"default_length": "medium",
"output_format": "text",
"language": "zh-CN",
"timeout": 30
}
# 模型配置
export SUMMARIZE_MODEL="openai/gpt-5.2"
export SUMMARIZE_MODEL="anthropic/claude-3.5-sonnet"
export SUMMARIZE_MODEL="google/gemini-3-flash-preview"
# 服务密钥
export FIRECRAWL_API_KEY="your_firecrawl_key" # 用于受阻网站
export APIFY_API_TOKEN="your_apify_token" # 用于YouTube
# 代理设置
export HTTP_PROXY="http://proxy.example.com:8080"
export HTTPS_PROXY="http://proxy.example.com:8080"
# 令牌限制
summarize https://example.com --max-output-tokens 1000
# 语言设置
summarize https://example.com --language zh-CN # 中文
summarize https://example.com --language en-US # 英文
summarize https://example.com --language ja-JP # 日文
# 超时设置
summarize https://example.com --timeout 60 # 60秒超时
# 详细输出
summarize https://example.com --verbose
# 调试模式
summarize https://example.com --debug
# 摘要多个新闻文章
summarize https://news.example.com/article1 --length short
summarize https://news.example.com/article2 --length medium
summarize https://news.example.com/article3 --length long
# 批量处理新闻URL
cat news_urls.txt | xargs -I {} summarize {} --length short
# 摘要PDF论文
summarize paper.pdf --length long
# 结构化输出论文信息
summarize research_paper.pdf --json | jq '.title, .authors, .summary'
# 批量处理论文文件夹
find ./papers -name "*.pdf" -exec summarize {} --length medium \;
# 摘要会议录音
summarize meeting_recording.mp3 --length medium
# 结合多个文件
summarize meeting_notes.docx meeting_recording.mp3 --output meeting_summary.txt
# 生成结构化会议纪要
summarize meeting.mp3 --json --output meeting.json
#!/bin/bash
# 每日新闻摘要脚本
NEWS_URLS=(
"https://news.example.com/headline1"
"https://news.example.com/headline2"
"https://news.example.com/headline3"
)
OUTPUT_FILE="daily_news_summary_$(date +%Y%m%d).txt"
echo "=== 每日新闻摘要 $(date) ===" > "$OUTPUT_FILE"
echo "" >> "$OUTPUT_FILE"
for url in "${NEWS_URLS[@]}"; do
echo "处理: $url"
summarize "$url" --length short >> "$OUTPUT_FILE"
echo "" >> "$OUTPUT_FILE"
echo "---" >> "$OUTPUT_FILE"
echo "" >> "$OUTPUT_FILE"
done
echo "摘要已保存到: $OUTPUT_FILE"
#!/bin/bash
# 研究资料摘要脚本
RESEARCH_DIR="./research_materials"
OUTPUT_DIR="./summaries"
mkdir -p "$OUTPUT_DIR"
# 处理PDF文件
find "$RESEARCH_DIR" -name "*.pdf" | while read pdf; do
basename=$(basename "$pdf" .pdf)
echo "处理PDF: $basename"
summarize "$pdf" --length medium --output "$OUTPUT_DIR/${basename}_summary.txt"
done
# 处理网页书签
if [ -f "$RESEARCH_DIR/bookmarks.txt" ]; then
while read url; do
if [ -n "$url" ]; then
domain=$(echo "$url" | sed -e 's|^[^/]*//||' -e 's|/.*$||')
safe_name=$(echo "$domain" | tr './' '_')
echo "处理网页: $domain"
summarize "$url" --length short --output "$OUTPUT_DIR/web_${safe_name}_summary.txt"
fi
done < "$RESEARCH_DIR/bookmarks.txt"
fi
1. 网站访问失败
# 使用Firecrawl服务
export FIRECRAWL_API_KEY="your_key"
summarize https://blocked-site.com --firecrawl always
# 使用代理
export HTTP_PROXY="http://proxy:8080"
summarize https://example.com
2. YouTube视频处理失败
# 设置Apify API密钥
export APIFY_API_TOKEN="your_apify_token"
summarize https://youtube.com/video --youtube auto
# 使用备用方法
summarize https://youtube.com/video --youtube always
3. 大文件处理超时
# 增加超时时间
summarize large_document.pdf --timeout 120
# 减少输出长度
summarize large_document.pdf --length short
# 分块处理(如果支持)
summarize large_document.pdf --chunk-size 1000
4. 内存不足
# 使用更小的模型
export SUMMARIZE_MODEL="openai/gpt-4o-mini"
summarize document.pdf
# 处理前压缩文件
pdftoppm -jpeg document.pdf | summarize -
# 启用详细输出
summarize https://example.com --verbose
# 启用调试模式
summarize https://example.com --debug
# 查看处理步骤
summarize https://example.com --trace
# 查看完整帮助
summarize --help
# 查看版本信息
summarize --version
# 列出支持的格式
summarize --formats
# 并行处理多个文件
parallel summarize {} --length short ::: *.pdf
# 使用缓存(如果支持)
summarize https://example.com --cache
# 预处理大文件
pdftotext large.pdf - | summarize - --length medium
提示: summarize工具非常适合快速获取内容概要,但对于需要深度分析或精确提取的任务,建议结合其他工具(如专门的PDF解析器、网页爬虫等)使用。