AGENT SKILL

daily-news-briefing

Universal daily news briefing generator. Aggregates and summarizes the latest news on ANY topic the user specifies (AI, finance, geopolitics, sports, tech, science, etc.) from multiple sources. Activates when user asks for today's news, daily briefing, or latest updates on any subject — e.g. '给我今天的AI资讯', '帮我整理一下美股新闻', '科技行业有什么新动态', '今日财经简报'.

License 许可
MIT
最近更新
2026-11-11
Stars
★ 0
Forks
0

01 前言 Front Matter

name
daily-news-briefing
description
Universal daily news briefing generator. Aggregates and summarizes the latest news on ANY topic the user specifies (AI, finance, geopolitics, sports, tech, science, etc.) from multiple sources. Activates when user asks for today's news, daily briefing, or latest updates on any subject — e.g. '给我今天的AI资讯', '帮我整理一下美股新闻', '科技行业有什么新动态', '今日财经简报'.

02 资源文件 5

03 SKILL.md 正文

? Universal Daily News Briefing

Aggregates the latest news on any topic from multiple sources and delivers concise summaries with direct links.

When to Use This Skill

Activate this skill when the user:

  • Asks for today's news / latest updates on any topic
  • Requests a daily briefing or news summary
  • Mentions wanting to know what's happening in a specific field or industry
  • Says: "给我今天的XX资讯" / "帮我整理XX新闻" / "XX有什么新动态"
  • Examples of supported topics: AI、科技、财经、美股、A股、加密货币、地缘政治、体育、科学、游戏、医疗...

Key principle: The topic is determined by the user's input, not hardcoded.


Workflow Overview

Step 0: Identify Topic & Scope from User Input
      ↓
Phase 1: Information Gathering
  ├─ Direct website fetching (3-5 topic-relevant news sites)
  └─ Web search with date filters
      ↓
Phase 2: Content Filtering
  ├─ Keep: Last 24-48 hours, significant developments
  └─ Remove: Duplicates, minor updates, off-topic content
      ↓
Phase 3: Categorization (topic-adaptive)
  └─ Organize into 4-6 relevant subcategories
      ↓
Phase 4: Output Formatting
  └─ Present with links and structure

Step 0: Identify Topic & Scope

Before gathering news, extract from the user's message:

Field How to determine
Topic Explicit mention ("AI", "财经", "美股") or inferred from context
Scope Global / regional / company-specific
Time range Default: last 24 hours; adjust if user specifies ("this week", "最近3天")
Language Match user's language; default: Chinese for Chinese users
Depth Brief / Standard (default) / Deep

If the topic is unclear, ask: "您想了解哪个领域的日报?(例如:AI、科技、财经、地缘政治等)"


Phase 1: Information Gathering

Step 1.1: Select Topic-Relevant Sources

Choose 3-5 primary sources based on the identified topic:

AI / Technology:

Finance / Markets (Global):

Finance / Markets (China / A股):

Crypto / Web3:

Geopolitics / World News:

Science:

Sports:

If the topic doesn't match a preset category above, search for the top 3-5 authoritative sources for that domain dynamically.

Step 1.2: Execute Web Search Queries

Use web_search with date-filtered queries:

Query Template (adapt keywords to the identified topic):

General:   "[TOPIC] news today" OR "[TOPIC] latest updates" after:[YESTERDAY_DATE]
Events:    "[TOPIC] announcement" OR "[TOPIC] breaking news" after:[YESTERDAY_DATE]
Analysis:  "[TOPIC] trends" OR "[TOPIC] analysis" after:[YESTERDAY_DATE]

For Chinese topics, also search in Chinese:

"[主题] 最新消息 今日"
"[主题] 新动态"

Best Practices:

  • Always use current date or yesterday's date in filters
  • Execute 2-3 queries across different angles
  • Prioritize results from last 24-48 hours

Step 1.3: Fetch Full Articles (Top 10-15)

For the most relevant stories found:

  • Fetch full article content using web_fetch
  • Ensures accurate summarization vs. just using snippets

Phase 2: Content Filtering

Filter Criteria

Keep:

  • News from last 24-48 hours (preferably today)
  • Major announcements, significant events, notable developments
  • Verified information from authoritative sources

Remove:

  • Duplicate stories (keep most comprehensive version)
  • Minor updates or marketing fluff
  • Content older than 3 days (unless highly significant)
  • Off-topic content

Phase 3: Categorization (Topic-Adaptive)

Dynamically define 4-6 subcategories based on the topic. Examples:

AI / Tech topic:

  • ? 重大发布 (Major Announcements)
  • ? 研究突破 (Research & Papers)
  • ? 行业动态 (Industry & Business)
  • ?️ 工具应用 (Tools & Applications)
  • ? 政策监管 (Policy & Regulation)

Finance / 财经 topic:

  • ? 市场行情 (Market Movements)
  • ? 公司动态 (Corporate News)
  • ? 宏观经济 (Macro & Economy)
  • ? 资金流向 (Capital & Investment)
  • ? 数据报告 (Data & Reports)

Geopolitics topic:

  • ⚡ 突发事件 (Breaking Events)
  • ? 外交动态 (Diplomacy)
  • ? 军事安全 (Military & Security)
  • ? 经济制裁 (Economic Measures)
  • ?️ 国内政治 (Domestic Politics)

Other topics: Derive relevant categories from the subject matter.


Phase 4: Output Formatting

Use this template, substituting [TOPIC] with the actual topic:

# ? [TOPIC] 日报 · [Current Date]

**主题**: [TOPIC]
**时间范围**: [Time range, e.g., 过去24小时]
**来源数量**: [X] 篇文章 · [Y] 个来源

---

## [Category Emoji] [Category Name]

### [Headline 1]

**摘要**: [One-sentence overview]

**要点**:

- [Key detail 1]
- [Key detail 2]
- [Key detail 3]

**影响/意义**: [Why this matters - 1 sentence]

? **来源**: [Publication Name] · [Publication Date]
? **链接**: [URL]

---

### [Headline 2]

[Same format]

---

## [Next Category]

[Same format]

---

## ? 今日要点

1. [Most important development]
2. [Second most important]
3. [Trend worth watching]

---

_生成时间: [Timestamp] | 下次更新: 明日同时_

Output language: Match the user's language (Chinese input → Chinese output, English input → English output).


Customization Options

After providing the initial briefing, offer:

Focus / Filter

  • "要我只看某类子板块吗?比如只看市场行情 / 只看研究论文"
  • Specific company or entity focus

Depth

  • Brief: Headlines + 2-3 bullets
  • Standard: Summaries + key points (default)
  • Deep: Include analysis and expert reactions

Time Range

  • Last 24 hours (default)
  • Last 3 days / Last week / Custom

Format

  • By category (default)
  • Chronological
  • By significance

Follow-up Interactions

User says Action
"Tell me more about [story X]" Fetch full article, provide detailed summary + analysis
"What are experts saying about [topic Y]?" Search for expert opinions and analysis pieces
"Change topic to [Z]" Re-run workflow with new topic
"Only show [subcategory]" Filter and reorganize output
"用中文/英文给我" Switch output language

Quality Standards

  • ✅ All links valid and accessible
  • ✅ No duplicate stories
  • ✅ All items timestamped (preferably today)
  • ✅ Summaries accurate (not hallucinated)
  • ✅ Links lead to original sources
  • ✅ Mix of sources (not all from one publication)
  • ✅ Categories match the topic (not generic)

Error Handling

  • web_fetch fails → Skip, try next source
  • Search returns no results → Expand date range or try different query
  • Content is paywalled → Use available excerpt, note limitation
  • Topic is ambiguous → Ask user to clarify before proceeding

Examples

Example 1: AI 资讯

User: "给我今天的AI资讯" Action: Topic=AI, run full workflow, present structured AI briefing


Example 2: 财经日报

User: "帮我整理一下今日财经新闻" Action: Topic=Finance/财经, select finance sources, categorize into market/macro/corporate


Example 3: 地缘政治

User: "美伊局势最新动态" Action: Topic=US-Iran/Geopolitics, search focused queries, present geopolitics briefing


Example 4: 自定义主题

User: "游戏行业今天有什么新动态?" Action: Topic=Gaming Industry, dynamically select gaming news sources, adapt categories accordingly


Example 5: 周报模式

User: "给我本周的科技新闻汇总" Action: Topic=Tech, time range=last 7 days, present weekly digest

04 同仓库更多 Skills

hellowKeyzj
hellowKeyzj/Matcha-claw

web-extract

获取社交/内容网站数据,Bilibili/知乎/微博/Reddit/HN/V2EX/雪球/微信读书/BOSS直聘/arXiv 等 60+ 站点的结构化数据,自动复用Chrome登录态,零风险。

★ 0
hellowKeyzj
hellowKeyzj/Matcha-claw

openclaw-maintenance

诊断和维护 Matcha 托管的本地 OpenClaw,按证据申请最小修复并核实恢复。Use when 用户说 OpenClaw 起不来、启动失败、运行异常,要求检查配置、日志、数据库或进行维护修复;不用于创建 Agent、通用系统管理或后台巡检。

★ 0
hellowKeyzj
hellowKeyzj/Matcha-claw

nuwa-skill

女娲造人:输入人名/主题/甚至只是模糊需求,自动深度调研→思维框架提炼→生成可运行的人物Skill。 两种入口:(1)明确人名→直接蒸馏 (2)模糊需求→诊断推荐→再蒸馏。 触发词:「造skill」「蒸馏XX」「女娲」「造人」「XX的思维方式」「做个XX视角」「更新XX的skill」。 模糊需求也触发:「我想提升决策质量」「有没有一种思维方式能帮我...」「我需要一个思维顾问」。

Skill与智能体开发
★ 0
hellowKeyzj
hellowKeyzj/Matcha-claw

multi-search-engine

凡是需要使用 web_search 搜索任何信息时必须优先激活。覆盖所有联网查询场景:新闻、资讯、事件、价格、人物、公司、热点等。结合多引擎覆盖(百度/微信/集思录等中文源),并自动读取原文确认内容完整性。适用于任何需要联网查询的场景。

★ 0
hellowKeyzj
hellowKeyzj/Matcha-claw

llm-wiki

通过 Matcha 共用 MCP 服务器 matcha 查询用户的 LLM Wiki 知识库,支持页面检索、正文读取、目录、知识图谱和源文件任务查看。仅在用户明确提到 LLM Wiki、我的 wiki、我的知识库 / knowledge base,或要求查询 wiki 页面、知识图谱、扫描 wiki 源文件时触发;不用于泛指笔记、本地文件、Obsidian、Notion、Apple Notes、Logseq 等其他工具或开放网络检索。自动使用用户在 Wiki 页或 Chat 输入框选择的全局当前库,不绑定会话。默认只读;导入、刷新、扫描、写入、删除与嵌入仅在用户明确要求时执行。

Skill与智能体开发
★ 0
hellowKeyzj
hellowKeyzj/Matcha-claw

app-plugin-create

制作可独立安装与分发的 Matcha 插件,按任务选择工程、工具、界面和验收路径。Use when 用户要求创建 reader 文件阅读器、connector 连接器或 application 应用,或继续开发尚未安装的插件;不用于读取已安装插件源码、修改或覆盖已安装插件。

★ 0