Token导航 LogoToken导航

Agent Skill · 研究检索

web-search-hub(网络搜索中心)

web-search-hub 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

web-search-hub 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为可写文件;许可证为 MIT-0,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 2,838。

总安装 53,611周安装 2,190GitHub Stars 1下载 17,345核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.1.3

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:web-search-hub(网络搜索中心)
来源仓库:https://github.com/anisafifi/web-search-hub
安装命令:
openclaw skills install web-search-hub
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 web-search-hub

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

当用户需要在网络上搜索信息、新闻、图像或视频时,请使用此技能。触发器包括:“搜索”、“查找有关信息”、“查找”、“最新消息”的请求,或任何需要当前网络内容的请求。也可用于研究任务、事实检查、查找视觉资源或收集最新新闻。需要从clawhub.ai 安装 OpenClawCLI。

当 Claude 的内置 web_search 工具更适合简单查询时,请勿使用。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install web-search-hub;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要可写文件,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

web-search-hub 怎么安装?

在终端执行 openclaw skills install web-search-hub。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

web-search-hub 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

web-search-hub 需要什么权限?

权限级别是可写文件。该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

web-search-hub 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

web-search-hub 用的人多吗?

公开数据是累计安装 53,611 次,近一周安装 2,190 次,下载 17,345 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

web-search-hub 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
web-search-hub

网络搜索中心 · 本页

53,6111可写文件MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
web-search-hub
description
Use this skill when users need to search the web for information, news, images, or videos. Triggers include: requests to \"search for\", \"find information about\", \"look up\", \"what's the latest on\", or any request requiring current web content. Also use for research tasks, fact-checking, finding visual resources, or gathering recent news. Requires OpenClawCLI installation from clawhub.ai. Do NOT use when Claude's built-in web_search tool is more appropriate for simple queries.
license
Proprietary

Web Search Hub

Search the web using DuckDuckGo's API. Supports web pages, news articles, images, and videos with customizable filtering and output formats.

⚠️ Prerequisite: Install OpenClawCLI (Windows, MacOS) and run pip install duckduckgo-search

Installation Best Practices:

  • If you encounter permission errors, use a virtual environment instead of system-wide installation
  • For virtual environment: python -m venv venv && source venv/bin/activate && pip install duckduckgo-search
  • Never use --break-system-packages as it can damage your system's Python installation

Quick Reference

TaskCommand
Basic web searchpython scripts/search.py "query"
Recent newspython scripts/search.py "topic" --type news --time-range w
Find imagespython scripts/search.py "subject" --type images
Find videospython scripts/search.py "tutorial" --type videos
Save resultspython scripts/search.py "query" --output file.txt
JSON outputpython scripts/search.py "query" --format json

Core Search Types

Web Search (Default)

Returns web pages with titles, URLs, and descriptions.

python scripts/search.py "quantum computing"
python scripts/search.py "python asyncio tutorial" --max-results 20

News Search

Returns articles with source, date, and summary.

python scripts/search.py "climate summit" --type news
python scripts/search.py "AI regulation" --type news --time-range d

Image Search

Returns images with URLs, thumbnails, dimensions, and source.

python scripts/search.py "mountain sunset" --type images
python scripts/search.py "abstract art" --type images --image-color Blue

Video Search

Returns videos with title, publisher, duration, date, and URL.

python scripts/search.py "cooking tutorial" --type videos
python scripts/search.py "documentary" --type videos --video-duration long

Essential Options

Result Count

--max-results N    # Default: 10, range: 1-unlimited

Examples:

python scripts/search.py "machine learning" --max-results 5   # Quick overview
python scripts/search.py "research topic" --max-results 30    # Comprehensive

Time Filtering

--time-range <d|w|m|y>
# d = past day
# w = past week  
# m = past month
# y = past year

Examples:

python scripts/search.py "tech news" --time-range d      # Today's news
python scripts/search.py "research papers" --time-range y # Recent publications

Region Selection

--region <code>    # Default: wt-wt (worldwide)

Common codes: us-en, uk-en, ca-en, au-en, de-de, fr-fr

Example:

python scripts/search.py "local events" --region us-en --type news

Safe Search

--safe-search <on|moderate|off>    # Default: moderate

Example:

python scripts/search.py "medical information" --safe-search on

Output Formats

Text (Default)

Clean, numbered results with URLs and descriptions.

python scripts/search.py "topic"

Output:

1. Page Title
   URL: https://example.com
   Description text here...

2. Next Result
   URL: https://example.com/page
   Description text...

Markdown

Formatted with headers, bold, and links.

python scripts/search.py "topic" --format markdown

Output:

## 1. Page Title

**URL:** https://example.com

Description text here...

JSON

Structured data for programmatic processing.

python scripts/search.py "topic" --format json

Output:

[
  {
    "title": "Page Title",
    "href": "https://example.com",
    "body": "Description text..."
  }
]

Save to File

--output <filepath>

Examples:

python scripts/search.py "AI trends" --output results.txt
python scripts/search.py "news" --type news --format markdown --output news.md
python scripts/search.py "data" --format json --output data.json

Image Search Filters

Size

--image-size <Small|Medium|Large|Wallpaper>

Example:

python scripts/search.py "landscape" --type images --image-size Large

Color

--image-color <color|Monochrome|Red|Orange|Yellow|Green|Blue|Purple|Pink|Brown|Black|Gray|Teal|White>

Example:

python scripts/search.py "abstract art" --type images --image-color Blue

Type

--image-type <photo|clipart|gif|transparent|line>

Example:

python scripts/search.py "icons" --type images --image-type transparent

Layout

--image-layout <Square|Tall|Wide>

Example:

python scripts/search.py "wallpaper" --type images --image-layout Wide

Video Search Filters

Duration

--video-duration <short|medium|long>

Example:

python scripts/search.py "recipe" --type videos --video-duration short

Resolution

--video-resolution <high|standard>

Example:

python scripts/search.py "tutorial" --type videos --video-resolution high

Common Workflows

Research a Topic

Gather comprehensive information across multiple search types:

# Web overview
python scripts/search.py "machine learning" --max-results 15 --output ml_web.txt

# Recent news
python scripts/search.py "machine learning" --type news --time-range m --output ml_news.txt

# Tutorial videos
python scripts/search.py "machine learning tutorial" --type videos --output ml_videos.txt

# Visual examples
python scripts/search.py "machine learning diagrams" --type images --max-results 20 --output ml_images.txt

Track Current Events

Monitor breaking news on specific topics:

python scripts/search.py "election results" --type news --time-range d --format markdown --output daily_news.md

Find Visual Resources

Search for images with specific requirements:

python scripts/search.py "data visualization" --type images --image-type photo --image-size Large --max-results 30 --output viz_images.txt

Fact-Check Information

Verify claims with recent sources:

python scripts/search.py "claim to verify" --time-range w --max-results 20 --output verification.txt

Market Research

Gather business intelligence:

python scripts/search.py "electric vehicle market 2025" --max-results 25 --output market_overview.txt
python scripts/search.py "EV industry" --type news --time-range m --output market_news.txt

Academic Research

Find scholarly resources:

python scripts/search.py "quantum entanglement" --time-range y --max-results 30 --format markdown --output research.md

Implementation Guidelines

When users request web searches, follow this approach:

1. Identify Intent

  • What content type? (web, news, images, videos)
  • How recent? (use --time-range for current info)
  • How many results? (adjust --max-results)
  • Any special filters? (size, color, duration, etc.)

2. Configure Search

python scripts/search.py "query" \
  --type <web|news|images|videos> \
  --max-results <N> \
  --time-range <d|w|m|y> \
  [additional filters]

3. Choose Format

  • Text: Quick reading, immediate review
  • Markdown: Documentation, formatted reports
  • JSON: Further processing, automation

4. Execute and Process

# Run search
python scripts/search.py "query" [options] --output results.txt

# Read results if needed
cat results.txt

# Extract URLs or combine multiple searches

Best Practices

Search Strategy

  1. Start specific - Use clear, targeted queries
  2. Use time filters - Apply --time-range for current topics
  3. Adjust result count - Start with 10-20, increase if needed
  4. Choose right type - News for current events, web for general info

Output Management

  1. Save important searches - Use --output to preserve results
  2. Use appropriate format - JSON for automation, markdown for docs
  3. Organize files - Create folders for multi-search research

API Usage

  1. Avoid rapid requests - Space out searches to prevent rate limiting
  2. Be efficient - Use filters to get better results with fewer searches
  3. Respect limits - Don't hammer the API unnecessarily

Troubleshooting

Installation Issues

"Missing required dependency"

# Standard installation
pip install duckduckgo-search

# If you get permission errors, use a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install duckduckgo-search

Important: Never use --break-system-packages flag as it can corrupt your system's Python installation. Always use virtual environments for isolated package management.

"OpenClawCLI not found"

  • Download from https://clawhub.ai/
  • Install for your OS (Windows/MacOS)
  • Verify installation: openclaw --version

Search Issues

"No results found"

  • Broaden search terms
  • Remove time filters
  • Try different query phrasing

"Timeout errors"

  • DuckDuckGo service may be temporarily unavailable
  • Wait a moment and retry
  • Check internet connection

"Unexpected results"

  • DuckDuckGo results differ from Google
  • Refine query with more specific terms
  • Try adding context to the query

Rate Limiting

"Too many requests"

  • Space out searches (wait 1-2 seconds between requests)
  • Reduce frequency if making automated searches
  • Consider batching queries instead of individual requests

Advanced Usage

Combining Multiple Searches

Build comprehensive research by combining search types:

# Create research folder
mkdir research

# Gather all content types
python scripts/search.py "topic" --max-results 20 --output research/web.txt
python scripts/search.py "topic" --type news --time-range m --output research/news.txt  
python scripts/search.py "topic" --type images --max-results 30 --output research/images.txt
python scripts/search.py "topic" --type videos --max-results 15 --output research/videos.txt

Programmatic Processing

Use JSON for automated workflows:

# Get JSON data
python scripts/search.py "research query" --format json --output data.json

# Process with custom script
python analyze_results.py data.json

Building Knowledge Bases

Create searchable documentation:

mkdir knowledge-base

# Search related topics
python scripts/search.py "topic1" --format markdown --output knowledge-base/topic1.md
python scripts/search.py "topic2" --format markdown --output knowledge-base/topic2.md
python scripts/search.py "topic3" --format markdown --output knowledge-base/topic3.md

Limitations

Search Capabilities

  • Results depend on DuckDuckGo's index (may differ from Google)
  • No advanced operators (no site:, filetype:, etc.)
  • Image/video results may be limited compared to web search
  • No control over ranking algorithms

Content Access

  • Cannot access paywalled content
  • Some sites may block DuckDuckGo crawler
  • Dynamic JavaScript content may not be indexed
  • Real-time data may have slight delays

API Constraints

  • Rate limiting applies to prevent abuse
  • No guaranteed uptime or availability
  • Results may vary by region and time
  • Some queries may be filtered for safety

Complete Command Reference

python scripts/search.py "<query>" [OPTIONS]

REQUIRED:
  query              Search query string (in quotes)

SEARCH TYPE:
  -t, --type         web|news|images|videos (default: web)

RESULTS:
  -n, --max-results  Maximum results (default: 10)
  --time-range       d|w|m|y (day, week, month, year)
  -r, --region       Region code (default: wt-wt)
  --safe-search      on|moderate|off (default: moderate)

OUTPUT:
  -f, --format       text|markdown|json (default: text)
  -o, --output       Save to file path

IMAGE FILTERS:
  --image-size       Small|Medium|Large|Wallpaper
  --image-color      color|Monochrome|Red|Orange|Yellow|Green|Blue|Purple|Pink|Brown|Black|Gray|Teal|White
  --image-type       photo|clipart|gif|transparent|line
  --image-layout     Square|Tall|Wide

VIDEO FILTERS:
  --video-duration   short|medium|long
  --video-resolution high|standard

HELP:
  --help             Show all options and usage examples

Examples by Use Case

Quick Searches

# Simple query
python scripts/search.py "python tutorials"

# Get more results
python scripts/search.py "python tutorials" --max-results 25

Current Events

# Today's news
python scripts/search.py "AI developments" --type news --time-range d

# This week's headlines
python scripts/search.py "technology" --type news --time-range w --max-results 30

Visual Content

# Find photos
python scripts/search.py "nature photography" --type images --image-type photo

# Specific color scheme
python scripts/search.py "office design" --type images --image-color Blue --image-size Large

# Transparent icons
python scripts/search.py "social media icons" --type images --image-type transparent

Video Content

# Short tutorials
python scripts/search.py "quick recipe" --type videos --video-duration short

# High-quality documentaries
python scripts/search.py "space documentary" --type videos --video-resolution high --video-duration long

Saved Research

# Create research report
python scripts/search.py "climate change solutions" --max-results 30 --format markdown --output climate_report.md

# Gather news archive
python scripts/search.py "tech industry" --type news --time-range m --format json --output tech_news.json

Support

For issues or questions:

  1. Check this documentation for solutions
  2. Run python scripts/search.py --help for command-line help
  3. Verify OpenClawCLI installation at https://clawhub.ai/
  4. Ensure duckduckgo-search library is installed

Key Resources:

  • OpenClawCLI: https://clawhub.ai/
  • DuckDuckGo Search Library: https://pypi.org/project/duckduckgo-search/

平台分布

OpenClaw

98.46%
按下载量换算17,078

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

未展示

权限和风险

可写文件

该 Skill 可能写入或修改本地文件,使用前需要确认目标目录和修改范围。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills