Chinese Game News Aggregation Skill

This skill allows the agent to fetch the latest articles, industry analysis, indie game spotlights, and general news from key Chinese game media portals.

🛠️ Usage Instructions

Run the Python script scripts/fetch_news.py located in this skill's directory (the same folder as this SKILL.md; typically installed at ~/.agents/skills/gamenews/). Requires Python 3 and network access — no third-party packages needed.

python <skill_dir>/scripts/fetch_news.py [options]

For example, when installed at the default location:

python ~/.agents/skills/gamenews/scripts/fetch_news.py --limit 5

Options:

--source <source_name>: Filter by a specific platform. Supported platforms:
- gcores: 机核网 (Console/Core gaming culture & podcasts)
- yystv: 游研社 (Game history, culture, and features)
- chuapp: 触乐网 (In-depth game features and indie news)
- gamersky: 游民星空 (General gaming news, reviews, and updates)
- 3dmgame: 3DMGame (PC gaming news and mods)
- gameres: GameRes游资网 (Game developer community & news)
- youxituoluo: 游戏陀螺 (Game entrepreneurship & industry trends)
- gamelook: 游戏大观 (B2B industry reports and news)
- indienova: Indienova (Independent game reviews and dev blogs)
- ign: IGN中国 (IGN Chinese edition)
- all (default): Aggregate from all platforms
--limit <number>: Number of articles to return per source (default is 5).
--days <number>: Only keep articles published within the last N days (0 = no limit, default). Articles with unparseable dates are kept.
--format <markdown|json>: Output format (default is markdown). Use json if you want to parse the structured data.
--query <keyword>: Only return articles whose title contains the specified keyword (case-insensitive).

Notes:

All dates are normalized to local time in the format YYYY-MM-DD HH:MM. Sources that only expose relative times (e.g. "3小时前") are resolved against the current time; a few may lack a precise date and keep their raw string.
HTML-based sources are sensitive to site redesigns. If a source returns zero items for a long time, its scraper may need updating — check with --source <name> --format json to see whether the source returned an error or simply no items.

💡 Example Commands

Get the latest 3 articles from all platforms (default Markdown):

python ~/.agents/skills/gamenews/scripts/fetch_news.py --limit 3

Search all news for articles mentioning "塞尔达" (Zelda):

python ~/.agents/skills/gamenews/scripts/fetch_news.py --query "塞尔达"

Fetch only Indie news from Indienova in JSON format:

python ~/.agents/skills/gamenews/scripts/fetch_news.py --source indienova --format json

Industry news from the last 2 days only:

python ~/.agents/skills/gamenews/scripts/fetch_news.py --source gamelook --days 2