tech-news

Original：🇨🇳 Chinese
Translated
4 scripts

Aggregate technology news and generate Chinese Markdown daily reports. Scrape sources including Hacker News, GitHub Trending, etc., translate titles and summaries, with optional image upload to R2. Use this tool when the user says phrases such as "generate today's tech news", "科技新闻", or "tech news".

15installs
Sourcefoundralab/my-skills
Added on

NPX Install

npx skill4agent add foundralab/my-skills tech-news

Tags

Translated version includes tags in frontmatter
news-aggregationmarkdown-daily-reportr2-image-uploadtech-news-scraperpython-utility

SKILL.md Content (Chinese)

View Translation Comparison →

Tech News Aggregation

Scrape popular tech news from multiple sources, translate to Chinese and generate a Markdown summary, with optional image upload to R2.

Use Cases

  • Generate daily tech news summaries
  • Need multi-source aggregation + Chinese summaries
  • Need Markdown reports with images

Prerequisites

  • Python 3.8+ (
    python3
    )
  • Translation API: 
    MINIMAX_API_KEY
    or 
    OPENAI_API_KEY
    (Minimax defaults to 
    MiniMax-M2.1-lightning
    )
  • Image upload (optional): 
    ~/.r2-upload.yml
    or 
    R2_UPLOAD_CONFIG
  • Network access to news sources and images
  • Optional performance parameters: 
    FETCH_WORKERS
    (default 4), 
    TRANSLATE_WORKERS
    (default 3)

Recommended Workflow

  1. Confirm date, sources, count, whether images are needed, and output path.
  2. Run 
    python3 scripts/generate.py ...
    .
  3. Review title translations, summary quality, duplicates, and image links.
  4. If image processing fails, regenerate with 
    --no-images
    or use 
    scripts/process_images.py
    to add missing images.

Quick Start

bash
python3 scripts/generate.py --date $(date +%F) --save ./news.md

Common Parameters

  • --sources <list>
    : Specify news sources (default: hackernews reddit-programming github-trending devto lobsters paperswithcode huggingface arxiv-ai)
  • --count <n>
    : Number of items to scrape per source (default: 15)
  • --limit <n>
    : Final curated count (default: 10)
  • --max-images <n>
    : Maximum number of images to process (default: 10)
  • --no-images
    : Disable image processing
  • --output-only
    : Output only Markdown

Classification Rules

  • AI & Machine Learning: ai, llm, gpt, claude, model
  • Development Tools & Open Source: rust, python, github, framework
  • Infrastructure & Cloud Native: cloud, aws, kubernetes, docker
  • Product & Design: product, design, ui, startup
  • Interesting News & Opinions: Others

Output Format

Adopt a fixed structure to ensure consistency:
markdown
# 📰 YYYY-MM-DD 科技早报

> 📊 **今日导读**
> 精选 10 条科技新闻
> 来源：Hacker News(4) | GitHub Trending(3) | Lobsters(3)

---

## 📋 文章速览

**AI与机器学习**：3 篇
1. 文章标题一
2. 文章标题二
...

---

## AI 与机器学习

### 1. 文章中文标题

📰 **Hacker News**

<img src="https://r2.example.com/image.jpg" width="100%">

**摘要**：中文摘要内容...

**核心要点**：
• 要点一
• 要点二
• 要点三

🔗 [阅读原文](https://example.com)

---
Complete templates and examples can be found in 
references/EXAMPLES.md
.

References

  • references/WORKFLOW.md
  • references/SOURCES.md
  • references/EXAMPLES.md