SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsnews-aggregator-skill

news-aggregator-skill

Writing & research

View repository

Updated 10-16 02:13

About this skill

News Aggregator Skill

Fetch real-time trending news from multiple sources.

Supported Data Sources

Data Source Identifier Type
Hacker News hackernews Tech/Startups
Weibo Hot Searches weibo Society/Entertainment
GitHub Trending github Open Source Projects
36Kr 36kr Tech/Business
Product Hunt producthunt Product Launches
V2EX v2ex Tech Community
Tencent News tencent General News
Wallstreetcn wallstreetcn Finance

Tool Usage

Basic Commands

uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py [参数]

Parameter Description

  • --source <源>: Specify data sources

    • Single source:hackernews, weibo, github, 36kr, producthunt, v2ex, tencent, wallstreetcn
    • Multiple sources (comma-separated):hackernews,github,producthunt
    • All sources:all

  • --limit <数量>: Maximum number of items returned per source (default: 10)

  • --keyword <关键词>: Keyword filtering (comma-separated)

    • Example:"AI,LLM,GPT"
    • Case-insensitive, supports word boundary matching

  • --deep: Enable deep fetching

    • Download and extract article body content (truncated to first 3000 characters)
    • Concurrent fetching to improve speed
    • Results will include the content field

Output Format

JSON array, each item contains:

  • source: Source name
  • title: Title
  • url: Link
  • heat: Popularity metric (points, reply count, star count, etc.)
  • time: Time information
  • content: Article content (only when using --deep when)

Usage Strategy

1. Global Scan (Broad Fetching)

Use cases: daily news roundup, comprehensive understanding of developments across fields

# 从所有源获取，每源 15 条，启用深度抓取
uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py --source all --limit 15 --deep

Note: A global scan returns about 120 items of data, and you need to perform semantic filtering and categorization based on user interests.

2. Single Data Source

Use cases: focus on a specific platform or field

# Hacker News 前 10 条
uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py --source hackernews --limit 10 --deep

# GitHub Trending 前 15 条
uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py --source github --limit 15 --deep

3. Keyword Search (Smart Expansion)

Key Rule: Automatically expand user keywords to cover the entire field

  • User says "AI" → Use:"AI,LLM,GPT,Claude,DeepSeek,Gemini,机器学习,RAG,Agent,大模型"
  • User says "frontend" → Use:"前端,React,Vue,Next.js,TypeScript,JavaScript,CSS,Vite"
  • User says "finance" → Use:"金融,股票,市场,经济,加密货币,比特币,黄金,A股"
# 示例：用户问 "有什么 AI 相关的新闻"
uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py \
  --source hackernews,github,36kr \
  --limit 20 \
  --keyword "AI,LLM,GPT,Claude,DeepSeek,Agent,大模型" \
  --deep

4. Precise Search

Only for very specific proper nouns

# 搜索 "DeepSeek" 相关新闻
uv run --directory .agents/skills/news-aggregator-skill python scripts/fetch_news.py --source all --limit 10 --keyword "DeepSeek" --deep

Output Specifications

Report Format Requirements

Language and Style：

  • Use Simplified Chinese
  • Adopt a magazine/newsletter style (such as The Economist or Morning Brew)
  • Professional, concise, engaging

Report Structure：

  1. Front Page Headlines(3-5 items)

    • The most important news across fields

  2. Tech and AI

    • A dedicated section for AI, LLM, and technology-related content

  3. Finance/Society

    • Other important categories (based on relevance)

Single news item format：

### 序号. [标题文本](原始URL)

**来源**：<数据源> | **时间**：<时间信息> | **热度**：<热度指标>

**核心要点**：一句话概括"所以呢？"

**深度解读**：
- 要点 1：为什么重要
- 要点 2：技术细节或背景
- 要点 3：影响和启示

Key rules：

  • ✅ The title must be a Markdown link：[OpenAI 发布 GPT-5](https://...)
  • ❌ Plain-text titles are prohibited：OpenAI 发布 GPT-5
  • The metadata line must include: source, time, popularity
  • During deep scanning, 2-3 interpretation points must be provided

Time filtering and smart supplementation

When the user specifies a time window (such as "past X hours") and results are scarce (< 5 items):

  1. Prioritize the user's window: first list items that strictly meet the time requirement
  2. Smart supplementation: if the list is too short, high-value/high-popularity items from a broader range must be included (such as the past 24 hours)
  3. Clearly label: clearly mark supplemented items (such as "⚠️ 18h ago", "🔥 24h popular")
  4. Value first: even if slightly outside the time window, prioritize showing SOTA, major releases, or high-popularity content

GitHub Trending special case：

  • Strictly return valid items from the scraped list (such as Top 10)
  • List all scraped items
  • Do not performsmart supplementation
  • Mustconduct an in-depth analysis of each project:
    • Core value: What problem does it solve? Why is it popular?
    • Inspired thinking: technical or product insights
    • Scenario tags: 3-5 keywords (such as #RAG #本地优先 #Rust）

Output file

  • Save location: the reports/ folder in the workspace root directory
  • File naming: with timestamp (such as hn_news_20260131_1430.md）
  • Full path example：/Users/dio/Documents/new_vault/reports/tech_news_20260131_1430.md
  • User display: present the full report content in the chat

Interactive menu

When the user says "news-aggregator-skill As You Wish"(or a similar "menu/help" trigger word):

  1. Read the templates.md file in the skill directory
  2. Show the user the list of available commands in the file
  3. Guide the user to select a number or copy a command to execute