Python-based Model Context Protocol (MCP) server that scrapes GeekNews articles and weekly news, caches the data for 24 h and exposes two tools (get_articles, get_weekly_news).
https://github.com/the0807/GeekNews-MCP-ServerStop switching between browser tabs to check Korean tech trends. The GeekNews MCP Server brings structured Korean technology news directly into your AI conversations and workflows.
If you're building products for Korean markets, tracking Asian tech trends, or just staying current with Korean innovation, you know the drill: bookmark GeekNews, remember to check it, manually parse what's relevant. This MCP server eliminates that friction entirely.
Pull the latest articles, weekly roundups, and trending discussions from GeekNews with simple function calls. Get structured data instead of hunting through HTML - titles, URLs, points, authors, timestamps, and comment counts all cleanly organized.
get_articles - Grab current articles by type (top, new, ask, show) with customizable limits. Perfect for daily trend monitoring or feeding your analysis workflows.
get_weekly_news - Access weekly roundups with full content and item lists. Ideal for comprehensive trend analysis or weekly report generation.
Both tools return structured data that your AI can immediately work with - no parsing, no cleanup, no manual formatting.
Here's what makes this server thoughtful: 24-hour intelligent caching with automatic background updates. The scheduler handles cache invalidation and refresh cycles, so you get fresh data without hammering GeekNews servers.
Cache files live in ~/.cache/geeknews-mcp, and the system automatically manages staleness checks and updates. You can call these tools frequently in your workflows without worrying about being a bad citizen.
Market Research Automation: Set up daily analysis of Korean tech trends by pulling top articles and feeding them to your analysis pipeline. Compare weekly roundups to spot emerging patterns.
Content Creation Workflows: Generate Korean tech summaries, trend reports, or market analysis by pulling structured data and having your AI synthesize insights.
Competitive Intelligence: Monitor specific topics or companies mentioned in Korean tech news without manual checking. Filter by article type to focus on discussions vs. announcements.
Development Planning: Stay informed about Korean developer tools, frameworks, and practices by accessing structured discussions and show posts.
Install via Smithery with one configuration block:
{
"mcpServers": {
"geeknews-mcp-server": {
"command": "npx",
"args": ["-y", "@smithery/cli@latest", "run", "@the0807/geeknews-mcp-server", "--key", "your-smithery-key"]
}
}
}
Or run locally with uv for development:
git clone https://github.com/the0807/GeekNews-MCP-Server
cd GeekNews-MCP-Server
uv sync && uv venv && source .venv/bin/activate
uv run mcp dev main.py
The server handles all the BeautifulSoup parsing, request management, and data structuring. You just call the tools and get clean JSON responses.
Instead of context-switching to check Korean tech news, integrate it directly into your AI workflows. Whether you're analyzing market trends, researching competitors, or staying current with Korean innovation, this server turns passive news consumption into active workflow integration.
The 24-hour cache means you can build this into daily automation without worrying about request limits or server load. The structured data output means your AI can immediately work with the content instead of you manually summarizing or organizing it.
For developers working with Korean markets or tracking Asian tech trends, this server transforms how you consume and analyze Korean technology news.