GeekNews-MCP-Server📰 GeekNews MCP Server

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

GeekNews MCP Server is a Model Context Protocol (MCP) server implemented in Python that retrieves articles from GeekNews. It utilizes BeautifulSoup for web scraping and features a caching mechanism to reduce server load by parsing data once a day and storing it in cache. The server provides tools to fetch articles and weekly news, with options to specify article types and return counts. Cached data is used to minimize requests to the GeekNews server, with cache data updated automatically every 24 hours.

Key features

get_articles tool for fetching articles
get_weekly_news tool for fetching weekly news
Cache management
Scheduler for periodic data updates
Support for specifying article types and return counts

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

Article Retrieval

Use the get_articles tool to fetch articles from GeekNews, specifying article types (top, new, ask, show) and return counts.

Weekly News Retrieval

Use the get_weekly_news tool to fetch weekly news from GeekNews, specifying a weekly news ID or retrieving the most recent one.

Pros & Cons

Pros

Reduces server load through caching
Provides easy-to-use tools for fetching articles and weekly news
Automatically updates cache once a day

Cons

Depends on the structure of the GeekNews website, which may change
Cache data is stored in the user's home directory

Reviews

No reviews yet. Be the first!

Q&A

What are the main limitations I should be aware of when using this MCP server?

The server relies on the current HTML structure of the GeekNews website, so any redesign may break the scraper until the parsing code is updated. Additionally, the cache resides on the local machine, so shared or multi‑user environments need separate caches.

Asked by Omar Haddad · May 24, 2025

Can I control which article types and how many articles are returned by the get_articles tool?

Yes. The `get_articles` tool lets you specify article types such as `top`, `new`, `ask`, or `show`, and you can set the number of articles to return. Each result includes title, URL, points, author, time, comment count, and rank.

Asked by Mateusz Wozniak · May 15, 2025

How does the caching system work and where is the cached data stored?

The server parses GeekNews once per day and saves the results in a cache located in the user's home directory under `.cache/geeknews_mcp`. Cached entries are valid for 24 hours; after that the scheduler automatically refreshes the data.

Asked by Idris Suleiman · May 10, 2025