Mcp Newsnow Server

Overview

The MCP NewsNow server is a lightweight, Model Context Protocol‑compatible service that aggregates trending news from more than fourteen popular Chinese and international platforms—including Coolapk, Zhihu, Weibo, Bilibili, Douyin, GitHub Trends, and others. By exposing a consistent set of tools (e.g., , , , and ) through the MCP interface, it allows AI assistants such as Claude to fetch real‑time news without embedding proprietary APIs or maintaining separate integrations for each source. This solves the problem of fragmented, platform‑specific news feeds and provides developers with a single, unified entry point for up-to-date content.

The server’s core value lies in its simplicity and flexibility. It relies on the open‑source Newsnow API, which already consolidates content from diverse sources, and adds a thin MCP wrapper that translates standard AI tool calls into HTTP requests. Developers can therefore focus on building conversational flows or data pipelines while the server handles source selection, language mapping, and rate‑limiting. The inclusion of fuzzy matching for both Chinese and English source names further reduces friction, enabling users to reference platforms in the language they prefer.

Key features include:

• Cross‑platform aggregation: Retrieve headlines from 14+ sources in one request, supporting both single and multi‑source queries.
• Language‑agnostic source identifiers: Accepts Chinese or English names (e.g., “知乎” or “zhihu”) and performs fuzzy matching to resolve ambiguities.
• Customizable API endpoint: The base URL for the underlying Newsnow service can be overridden via an environment variable (), making it easy to switch between production and staging deployments.
• Toolset variety: From fetching a single source’s latest news to listing all available sources, the server offers a comprehensive toolkit that can be invoked directly from an AI conversation or scripted in a backend workflow.

In practice, the MCP NewsNow server is ideal for applications that require real‑time trend analysis, content recommendation engines, or dynamic news widgets. For example, a chatbot could answer “What’s trending on Weibo right now?” by calling , or a data‑science pipeline could batch multiple sources with to feed a sentiment analysis model. Because the server follows MCP conventions, it plugs seamlessly into any AI workflow that supports tool invocation, whether in a desktop assistant, a web‑based chatbot, or an automated reporting system.

Unique advantages of this implementation are its minimal operational footprint and open‑source nature. The server itself is a single Python package that can be installed from PyPI, and it can run in isolated environments or be integrated into existing Claude Desktop configurations. Its reliance on the already‑established Newsnow API eliminates duplicate effort, while the MCP wrapper guarantees that developers can harness the power of AI assistants without needing to write custom adapters for each news platform.