Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计通过

strd-content-summarizerstrd 内容摘要器

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

20,955

周安装

856

GitHub Stars

1

下载量

6,711
OpenClaw

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

MIT-0

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:strd-content-summarizer(strd 内容摘要器)
来源仓库:https://github.com/kryzl19/strd-content-summarizer
安装命令:
openclaw skills install strd-content-summarizer
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install strd-content-summarizer

简介

strd-content-summarizer 用于提取 URL 内容并生成结构化摘要,支持 AI 增强提示模板。

  • 适合整理文档、README 或网页信息,提炼关键结构与术语统一。
  • 可辅助生成可读性强的说明文稿,但应保留已有事实,避免虚构结论。
  • 涉及对外文案时需注意语气控制,防止过度营销或夸大能力描述。
  • 安装前请核实权限与维护状态,确认是否触发网络访问或文件读写。

SKILL.md

name
content-summarizer
description
Fetch any URL and produce a structured content summary with extractive summarization, AI enhancement prompts, and structured output templates. Extracts clean text from HTML, identifies key sentences, generates metadata, and provides ready-to-use prompts for AI summarization. Perfect for content repurposing, research pipelines, and newsletter workflows.
metadata

Content Summarizer

Transform any URL or raw text into structured, actionable content summaries using extractive summarization + AI enhancement prompts.

What This Skill Does

  1. Fetches any URL and extracts clean text (removes HTML, nav, scripts, ads)
  2. Extracts key sentences algorithmically (extractive summarization — no API needed)
  3. Generates metadata (title, word count, reading time)
  4. Outputs a structured markdown template + AI enhancement prompt ready for any LLM

The AI enhancement prompt makes the extracted content easy to summarize with MiniMax, GPT-4o, or Claude.

Quick Start

# Summarize a URL (outputs to terminal + /tmp/summaries/)
./scripts/url-to-summary.sh https://example.com/article

# Summarize text directly (from argument or pipe)
./scripts/summarize.sh "Your article text here..."
cat article.txt | ./scripts/summarize.sh

# Custom model for AI prompt
./scripts/url-to-summary.sh https://example.com --model openai

Scripts

url-to-summary.sh — Full Summary from URL

./scripts/url-to-summary.sh <url> [--model minimax|openai|anthropic]

Fetches URL, extracts clean text, produces:

  • Extracted key sentences (extractive summarization)
  • Structured markdown template (summary, tweet hook, key takeaways)
  • AI enhancement prompt — copy-paste into any LLM for full summarization
  • Metadata (title, word count, reading time)

Output saved to /tmp/summaries/summary-TIMESTAMP.md

summarize.sh — Summarize Text Directly

./scripts/summarize.sh "Your text..."
cat file.txt | ./scripts/summarize.sh

Same output as url-to-summary.sh but for raw text input.

Output Format

# Content Summary

**Source:** [url]
**Title:** [extracted or unknown]
**Stats:** ~1,200 words | ~6 min read

## Key Sentences (Extractive Summary)
[Top 15-20 most meaningful sentences from the text]

## Structured Summary Template
[TODO fields for: what, why, key points]

## Tweet Hook
[TODO: Compelling one-liner]

## Key Takeaways
1. [TODO]
2. [TODO]
3. [TODO]

## AI Enhancement Prompt
[Ready-to-use prompt with extracted text embedded]

Environment Variables

VariableDefaultDescription
OUTPUT_DIR/tmp/summariesWhere to save summary files
MODELminimaxAI model for enhancement prompt

Use Cases

  • Content repurposing: Fetch articles, extract key points, feed into content pipeline
  • Research pipeline: Batch summarize 20 URLs, collect key points, synthesize insights
  • Newsletter workflow: Fetch sources, generate structured summaries, assemble newsletter
  • Quick reading: Extract key sentences from long articles for rapid triage

Notes

  • Uses extractive summarization (algorithmically selects important sentences) — no API key needed
  • AI enhancement prompt is the value-add: paste into MiniMax/GPT/Claude for full abstractive summary
  • HTML extraction removes scripts, styles, nav, footer, ads — focuses on article body
  • Limited to 15,000 characters of extracted text to keep context manageable

Output Format

## Summary

[2-3 paragraph summary of the content]

## Tweet Hook
> [One compelling, shareable sentence that captures the essence]

## Key Takeaways
1. [First key point]
2. [Second key point]
3. [Third key point]

---
*Source: [URL] | ~X min read | ~Y words*

Environment Variables

VariableRequiredDefaultDescription
SUMMARIZER_MODELNoautoLLM model for summarization (or let the agent auto-select)
MAX_TOKENSNo500Max tokens in summary output

Usage Examples

# Summarize a blog post
./scripts/url-to-summary.sh https://yourcompetitor.com/blog/industry-trends

# Process a research paper
./scripts/summarize.sh "$(curl -sL https://arxiv.org/pdf/2301.00001.pdf | head -c 10000)"

# Quick key points for a meeting
./scripts/key-points.sh https://news.ycombinator.com/item?id=12345678

# Use with OpenClaw: feed the output directly into your workflow
SUMMARY=$(./scripts/url-to-summary.sh https://example.com/article)
echo "$SUMMARY" | ./scripts/key-points.sh

Tips

  • For better results, provide clean article URLs (avoid paywalls and heavy JS pages)
  • Use MAX_TOKENS=1000 for longer, more detailed summaries
  • Chain with other skills: competitor-tracker alerts → content-summarizer → social posting
  • The tweet hook is designed to be engaging, not just a truncated sentence
  • Key points should be actionable insights, not just topic labels

Dependencies

  • curl — for fetching URLs
  • OpenClaw agent with web_fetch tool for content extraction (used internally by the agent)

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

98.85%
按下载量换算6,634

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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