← 返回 Skills 市场
372
总下载
0
收藏
4
当前安装
1
版本数
在 OpenClaw 中安装
/install news-search
功能描述
专门用于查询各大搜索网站的新闻结果。支持百度新闻、今日头条、Bing新闻、Google新闻和新浪新闻。可以同时聚合多个来源的结果,并进行去重处理。
使用说明 (SKILL.md)
新闻查询技能 (News Search Skill)
该技能使用 Playwright 无头浏览器访问各大搜索引擎的新闻频道,提取最新的新闻标题、链接、来源和发布时间。
支持的新闻源
- 百度新闻 (baidu):
https://www.baidu.com/s?tn=news&word={keyword} - 今日头条 (toutiao):
https://so.toutiao.com/search?keyword={keyword}&pd=news - Bing 新闻 (bing):
https://www.bing.com/news/search?q={keyword} - Google 新闻 (google):
https://news.google.com/search?q={keyword} - 新浪新闻 (sina):
https://search.sina.com.cn/?q={keyword}&c=news
使用方法
进入脚本目录并运行 news-search.js:
cd /Users/wuwei/.openclaw/workspace/skills/news-search/scripts
node news-search.js \x3Ckeyword> [options]
参数说明
keyword: 搜索关键词(必填)。
选项说明
--engine=NAME: 指定新闻引擎。可选值:baidu,toutiao,bing,google,sina,all(默认)。--max=N: 每个引擎返回的最大结果数(默认 10)。--format=md|json: 输出格式(默认 md)。
示例
# 搜索所有来源的关于 "人工智能" 的新闻
node news-search.js "人工智能"
# 仅使用百度新闻搜索 "特斯拉财报"
node news-search.js "特斯拉财报" --engine=baidu
# 获取 JSON 格式的结果并限制数量
node news-search.js "A股今日走势" --engine=toutiao --max=5 --format=json
输出示例 (Markdown)
============================================================
🔍 "人工智能" 新闻搜索结果(共 15 条)
============================================================
1. 2025年人工智能十大趋势发布
🔗 https://news.example.com/article1
📌 来源: 新华网 | 渠道: 百度新闻 | 时间: 2小时前
2. OpenAI 发布最新大模型预览版
🔗 https://tech.example.com/article2
📌 来源: 腾讯科技 | 渠道: 今日头条新闻 + Bing News
...
依赖项
playwright: 浏览器自动化工具。playwright-extra: Playwright 扩展,支持插件。puppeteer-extra-plugin-stealth: 增强反爬虫绕过能力。
安装
cd /Users/wuwei/.openclaw/workspace/skills/news-search/scripts
npm install
安全使用建议
This skill is internally consistent for scraping news search results, but consider the following before installing: (1) npm install will download Playwright and browser binaries (large, network I/O); (2) the script uses a stealth plugin and runs Chromium with --no-sandbox — avoid running it as root and consider running inside an isolated environment (container/VM) to reduce risk; (3) scraping search engines may violate service terms or triggers anti-bot defenses — use rate-limiting and respect robots.txt and legal constraints; (4) as a good practice, inspect and pin dependency versions before npm install and run the code in a sandbox if you are unsure.
功能分析
Type: OpenClaw Skill
Name: news-search
Version: 1.0.0
The news-search skill bundle is a legitimate tool for aggregating news from multiple search engines (Baidu, Bing, Google, etc.) using Playwright. The code in news-search.js implements standard web scraping logic with anti-bot measures (stealth plugin) and lacks any indicators of data exfiltration, unauthorized execution, or malicious intent.
能力评估
Purpose & Capability
Name/description say 'news search' and the package contains a Playwright-based scraper that targets the listed news search URLs; dependencies (playwright, playwright-extra, stealth plugin) are consistent with web scraping.
Instruction Scope
SKILL.md instructs running the provided Node script and npm install; it does not request reading unrelated local files or environment variables. Minor oddity: SKILL.md uses a hard-coded local path example (/Users/wuwei/...), which is just an example but may confuse users.
Install Mechanism
There is no custom install spec, but package.json declares common npm packages. Installing will pull Playwright and associated browser binaries (large downloads). All packages resolve via npm (package-lock references a mirror) — no arbitrary URL downloads or extract-from-unknown-host entries were found.
Credentials
The skill requires no environment variables, credentials, or config paths. No secrets are requested or accessed by the code.
Persistence & Privilege
Skill is not marked always:true and does not modify other skills or system configs; it runs as a one-off script, so no elevated persistence is requested.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install news-search - 安装完成后,直接呼叫该 Skill 的名称或使用
/news-search触发 - 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 the news-search skill.
- Supports searching news from Baidu, Toutiao, Bing, Google, and Sina.
- Aggregates results from multiple sources and removes duplicates.
- Offers options to specify news engines, result limits, and output formats (Markdown or JSON).
- Uses Playwright and related libraries for web scraping and anti-bot measures.
元数据
常见问题
news-search 是什么?
专门用于查询各大搜索网站的新闻结果。支持百度新闻、今日头条、Bing新闻、Google新闻和新浪新闻。可以同时聚合多个来源的结果,并进行去重处理。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 372 次。
如何安装 news-search?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install news-search」即可一键安装,无需额外配置。
news-search 是免费的吗?
是的,news-search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
news-search 支持哪些平台?
news-search 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 news-search?
由 wei.wu(@dlutwuwei)开发并维护,当前版本 v1.0.0。
推荐 Skills