About angheljf/nyt
NYTimes Article Search MCP Server is a TypeScript-based server that allows users to search for New York Times articles published within the last 30 days using a keyword. It implements the Model Context Protocol (MCP) and is designed to be used with AI assistants like Claude Desktop. The server provides a single tool, search_articles, which takes a keyword as input and returns a list of matching articles including title, abstract, URL, published date, and author. It integrates with the official New York Times API and requires an API key.
Key Features
Provides a single `search_articles` tool for keyword-based article search
Searches articles from the last 30 days
Returns title, abstract, URL, published date, and author for each article
Integrates with the New York Times API
Built with TypeScript
Designed for use with Claude Desktop and other MCP-compatible clients
Open source under MIT license
Pros & Cons
Pros
- Simple, focused tool for searching NYTimes articles
- Returns structured, detailed article information (title, abstract, URL, date, author)
- Easy installation with Claude Desktop via Smithery or manual configuration
- Open source with permissive MIT license
- Active development with community stars and forks on GitHub
Cons
- Only searches articles from the last 30 days, no historical search
- Requires a separate NYTimes API key to function
- Limited to single keyword input; no advanced query filters or boolean operators
- No built-in support for multiple keywords or batch queries
- Dependent on NYTimes API availability and rate limits
Best For
Finding recent news articles by topicIntegrating NYTimes article search into conversational AI assistantsResearch and content curation focused on current eventsDemonstrating MCP tool integration for custom AI workflows
FAQ
What tools does this MCP server provide?
It provides a single tool, `search_articles`, which searches New York Times articles from the last 30 days based on a keyword.
How do I install this server for Claude Desktop?
You can install it automatically via Smithery using `npx -y @smithery/cli install nyt --client claude`, or manually by adding the server configuration to your Claude Desktop config file as shown in the README.
What information is returned for each article?
Each article result includes the title, abstract, URL, published date, and author.
Do I need an API key?
Yes, you need a New York Times API key. Set it as the environment variable NYTIMES_API_KEY in your .env file or in the MCP server configuration.
Is this project open source?
Yes, it is open source under the MIT License.