← Back to Skills Marketplace
42
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install sina-daily-news
Description
抓取新浪财经 7×24 实时新闻流,自动分类生成结构化日报。每类新闻按时间线排列,标注来源和阅读量。适合每日早间财经/政治/科技新闻速览。
README (SKILL.md)
新浪每日新闻抓取
从新浪财经 7×24 实时新闻流自动抓取、分类、生成每日新闻报告。
功能
- 抓取
https://finance.sina.com.cn/7x24/?tag=0的 7×24 实时新闻 - 自动分类:国际政治军事 / 中国政治外交 / 中国财经股市 / 科技与AI / 产业与制造 / 社会热点
- 提取每条新闻的时间、标题、正文、来源、阅读量
- 生成结构化 Markdown 报告
使用方法
快速抓取
使用 OpenClaw 的 web_fetch 工具抓取页面,然后将内容传入解析脚本:
# 1. 用 web_fetch 抓取 https://finance.sina.com.cn/7x24/?tag=0 (maxChars: 15000)
# 2. 将抓取内容传入脚本
python scripts/sina_daily_news.py
Cron 定时任务
建议配置 OpenClaw Cron 任务,每天早晨 8:00 自动执行:
{
"name": "新浪每日新闻日报",
"schedule": { "kind": "cron", "expr": "0 8 * * *", "tz": "Asia/Shanghai" },
"sessionTarget": "isolated",
"payload": {
"kind": "agentTurn",
"message": "1. web_fetch 抓取 https://finance.sina.com.cn/7x24/?tag=0 (maxChars: 15000)\
2. 将内容传入 scripts/sina_daily_news.py\
3. 输出报告到聊天 + 保存到 memory/sina_news_YYYY-MM-DD.md",
"timeoutSeconds": 300
},
"delivery": { "mode": "announce", "channel": "last" }
}
脚本参数
| 参数 | 说明 |
|---|---|
| stdin | 传入抓取到的新闻文本(推荐方式) |
| argv[1] | 直接传入新闻文本字符串 |
输出格式
# 📰 新浪每日新闻日报 — YYYY-MM-DD
> 数据源:新浪财经 7×24 实时新闻
## 一、国际政治军事(N条)
- **[HH:MM:SS] 标题** | 来源 | 阅读 X万
正文摘要...
## 二、中国财经股市(N条)
...
分类规则
| 类别 | 关键词示例 |
|---|---|
| 国际政治军事 | 伊朗、美国、以色列、北约、俄罗斯、外交、总统 |
| 中国政治外交 | 中国、国务院、外交部、新华社、香港、台湾 |
| 中国财经股市 | A股、沪指、央行、利率、美联储、IPO、房地产 |
| 科技与AI | AI、大模型、芯片、华为、京东方、字节跳动 |
| 产业与制造 | 汽车、飞机、船舶、轨道交通、电站 |
| 社会热点 | 交通事故、食品安全、天气预警、邮轮 |
文件结构
sina-daily-news/
├── SKILL.md # 本文件
├── scripts/
│ └── sina_daily_news.py # 新闻解析脚本
└── package.json # ClawHub 元数据
故障排除
| 问题 | 解决 |
|---|---|
| web_fetch 403 | 改用 browser 工具打开页面并截图 |
| 解析脚本无输出 | 检查抓取内容是否包含有效新闻格式 |
| 分类不准确 | 编辑脚本中的 CATEGORIES 关键词列表 |
与其他新闻源互补
本脚本抓取新浪财经一手新闻流,建议与 web_search 多引擎搜索的新闻日报配合使用,形成双源交叉验证。
Usage Guidance
Before installing, note that the optional cron example would run daily and may save generated news reports to memory. Use it only if you are comfortable with scheduled public-web fetching from Sina Finance and stored daily report files.
Capability Assessment
Purpose & Capability
The stated purpose is daily Sina Finance news collection and categorization; the script only parses supplied text and prints a Markdown report.
Instruction Scope
Runtime instructions are limited to fetching one public URL, passing the content to the parser, and optionally producing a report.
Install Mechanism
The package contains only a skill document, simple package metadata, and one Python script with no declared dependencies or install hooks.
Credentials
The requested web access is proportionate to the news-reporting purpose, and the script does not read credentials, local files, browser profiles, or private data.
Persistence & Privilege
The documentation suggests an optional daily cron task and saving output under memory/sina_news_YYYY-MM-DD.md; this is disclosed, limited to generated reports, and user-configured.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install sina-daily-news - After installation, invoke the skill by name or use
/sina-daily-news - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
初次发布
Metadata
Frequently Asked Questions
What is 新浪每日新闻?
抓取新浪财经 7×24 实时新闻流,自动分类生成结构化日报。每类新闻按时间线排列,标注来源和阅读量。适合每日早间财经/政治/科技新闻速览。 It is an AI Agent Skill for Claude Code / OpenClaw, with 42 downloads so far.
How do I install 新浪每日新闻?
Run "/install sina-daily-news" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 新浪每日新闻 free?
Yes, 新浪每日新闻 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 新浪每日新闻 support?
新浪每日新闻 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 新浪每日新闻?
It is built and maintained by paudyyin (@paudyyin); the current version is v1.0.0.
More Skills