← Back to Skills Marketplace
92
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install china-briefing
Description
多尺度中文资讯简报生成。从一条街道到全球，任意地理范围均可生成。 从主流媒体（Reuters/BBC/新华社）、街道公众号、社交平台（微博/小红书/抖音） 获取资讯，结构化输出。 使用场景：用户要求"简报"、"资讯"、"早报"、"晚报"、"周报"、"XX简报"、 "XX资讯"、"本地新闻"、"国际新闻"、"街道新...
README (SKILL.md)
China Briefing — 多尺度中文资讯简报
核心能力
任意尺度，一处生成：
- 全球 → 区域（亚太/欧洲/中东） → 国家 → 省 → 市 → 区 → 街道→ 社区
- 每层自动选择匹配的搜索源
简报结构（固定，不省略）
### 🌍 核心结论
[1-2句总结本期最重要的信号/趋势/结论]
### 📌 {日期} {地点} 速报
**头条1 + 头条2 + 头条3**（最多3个，每个含嘟嘟点评）
### 📊 分类热点
[2-4个主题分类，每类2-4条]
### 📈 数据看板
[可选：关键数据表格]
### 🎯 行动建议
[优先级表格：高/中/低]
搜索源库（按尺度选源）
全球尺度（≥5种来源）
| 媒体 | 定位 | 获取方式 |
|---|---|---|
| Reuters | 事实最快 | web_search |
| BBC中文 | 国际视角 | web_search |
| CNN | 突发新闻 | web_search |
| Al Jazeera | 中东独特 | web_search |
| NYT中文 | 深度分析 | web_fetch |
| FT中文 | 经济深度 | web_fetch |
| 日经中文 | 亚洲经济 | web_fetch |
| DW德国之声 | 欧洲+中东 | web_search |
全国尺度
| 媒体 | 获取方式 |
|---|---|
| 新华社 | web_search |
| 央视新闻 | web_search |
| 环球时报 | web_search |
| 观察者网 | web_search |
| 澎湃新闻 | web_search |
省市尺度
| 来源 | 搜索策略 |
|---|---|
| 本地官媒 | "{省/市} 2026 最新动态" |
| 市政府网 | "{省/市} 政府 新闻" |
| 省/市人大 | "{省/市} 人大 会议" |
| 省级日报 | "{省}日报 最新" |
区县街道尺度（核心差异化）
| 来源 | 搜索策略 |
|---|---|
| 政府官网 | "{区/街道名} 2026 最新" |
| 政务公众号 | "{区/街道名} 公众号 民生" |
| 人大/政务 | "{区} 人大 两会 实事" |
| 本地社区 | "{区/街道} 社区 改造 电梯" |
| 本地媒体 | "{市} 新闻 {区}" |
| 本地生活 | "{街道} 公园 学校 医院" |
社交平台
| 平台 | 搜索策略 |
|---|---|
| 微博热搜 | "{关键词} 微博热搜" |
| 小红书 | "{关键词} 小红书" |
| 抖音 | "{关键词} 抖音" |
| Twitter/X | "{关键词} site:twitter.com" |
| "{关键词} site:reddit.com" |
使用流程
用户说 "国际简报"
- 用全球尺度源：Reuters + BBC + CNN + 半岛 + NYT + 新浪国际
- 搜索词："international news latest" + "国际新闻 最新"
- 输出 5-8 来源，标注出处
用户说 "江川路街道简报"
- 用区县街道尺度源
- 搜索词："江川路街道 2026 民生" + "江川路街道 实事" + "闵行区 江川路 新闻"
- 输出 3-5 来源，标注出处
用户说 "上海简报"
- 省市尺度源
- 搜索词："上海 2026 最新动态" + "上海 经济 民生 政策"
- 输出 4-6 来源
用户说 "AI行业简报"
- 行业垂直源 + 社区
- 搜索词："AI news latest 2026" + "AI 大模型 最新"
- 来源：行业媒体 + 社区 + 研报
评论公式
每条头条必须写"嘟嘟点评"，遵守：
❌ 只是描述："XX报道了XXX"
✅ 加背景："XX做XXX，结合之前的YYY，说明ZZZ"
✅ 给观点："这意味着...我看好/担心..."
✅ 连点："这和上个月的XX配对，说明趋势在..."
✅ 对读者："对我们来说，意味着..."
信息筛选
| 问题 | 不通过 → |
|---|---|
| 读者会分享给朋友？ | 去掉 |
| 有实质内容还是标题党？ | 去掉 |
| 来源可靠？ | 换来源 |
| 和读者相关？ | 标注为什么 |
| 能加独特观点？ | 加 |
来源多样性规则
| 尺度 | 最少来源数 | 必须包含 |
|---|---|---|
| 全球 | 5+ | Reuters + 1家中国官媒 + 1家社区 |
| 全国 | 4+ | 至少2家不同立场 |
| 省市 | 3+ | 官方 + 本地 |
| 区县/街道 | 2+ | 政府官网 + 至少1个来源 |
| 行业 | 3+ | 行业媒体 + 社区 |
每条信息必须标注来源。
字数控制
- 核心结论：50-100字
- 每条头条：150-300字（含点评）
- 分类热点：每条30-80字
- 总篇幅：800-2500字
References
- 搜索模板 →
references/search-templates.md - 来源库 →
references/source-library.md
Usage Guidance
This skill appears coherent and safe from an internal‑consistency perspective, but consider the following before installing: (1) It scrapes public web and social sources via web_search/web_fetch — confirm that your environment's fetch tools obey site terms, rate limits, and robot policies. (2) No credentials are requested, but some sources (WeChat/Weibo) may return limited results without logins; expect degraded coverage rather than secret exfiltration. (3) If you plan to brief very small communities or include personally identifiable information (PII), be mindful of privacy implications and local regulations. (4) Review source selection rules in references/source-library.md if you want to restrict or prioritize certain media (for political balance or compliance). (5) Run a few non‑sensitive queries to validate outputs and make sure the agent records and cites sources as required by the SKILL.md.
Capability Analysis
Type: OpenClaw Skill
Name: china-briefing
Version: 1.0.0
The 'china-briefing' skill bundle is a well-structured tool designed for generating multi-scale news reports. The Python script (scripts/briefing_helper.py) uses standard logic to categorize user queries into geographic or industrial scales, and the instructions in SKILL.md provide clear, task-aligned guidance for the AI agent to use web search tools. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description match the implementation: SKILL.md and scripts/briefing_helper.py only implement scale detection and search/query templates and call the platform web_search/web_fetch tools. No unrelated environment variables, binaries, or external services are requested.
Instruction Scope
Runtime instructions limit behavior to selecting sources, constructing search queries, fetching pages via OpenClaw web_search/web_fetch, filtering and annotating results. The instructions do not direct reading of local files, arbitrary env vars, or transmitting data to unexpected external endpoints.
Install Mechanism
This is instruction‑only with a small helper script included; there is no install spec, no downloads from third‑party URLs, and no extracted archives. The helper script is self‑contained Python using only standard library.
Credentials
No environment variables, credentials, or config paths are required. The claimed external dependency (OpenClaw web_search + web_fetch) is appropriate for a web‑based news aggregator; README explicitly states no API keys are required.
Persistence & Privilege
always:false (default) and no requests to modify other skills or agent config. The skill can be invoked autonomously (platform default) which is expected for skills of this type and is not combined with other concerning privileges.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install china-briefing - After installation, invoke the skill by name or use
/china-briefing - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
多尺度中文资讯简报。从一条街道到全球，21+权威来源（主流媒体、政府官网、社交平台）自动抓取，固定结构输出（核心结论→头条3条含嘟嘟点评→分类热点→数据看板→行动建议）。支持任意地理范围：全球、国家、省、市、区、街道。无需API Key。
Metadata
Frequently Asked Questions
What is china-briefing-资讯简报?
多尺度中文资讯简报生成。从一条街道到全球，任意地理范围均可生成。 从主流媒体（Reuters/BBC/新华社）、街道公众号、社交平台（微博/小红书/抖音） 获取资讯，结构化输出。 使用场景：用户要求"简报"、"资讯"、"早报"、"晚报"、"周报"、"XX简报"、 "XX资讯"、"本地新闻"、"国际新闻"、"街道新... It is an AI Agent Skill for Claude Code / OpenClaw, with 92 downloads so far.
How do I install china-briefing-资讯简报?
Run "/install china-briefing" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is china-briefing-资讯简报 free?
Yes, china-briefing-资讯简报 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does china-briefing-资讯简报 support?
china-briefing-资讯简报 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created china-briefing-资讯简报?
It is built and maintained by Shine8592 (@shine8592); the current version is v1.0.0.
More Skills