APITube Documentation

Appearance

Sign inGet started

Using News API with AI ​

APITube is built to be used by AI assistants and agents, not just by code. Ask for news in plain language from Cursor, Claude, or ChatGPT and get back structured, filtered results from 500,000+ sources — no manual request building required.

This section brings together everything you need to use the News API with AI tools.

Start here ​

  • MCP Server — connect any MCP-compatible assistant to the hosted server at https://mcp.apitube.io/. No install, just an API key.
  • Code Editors Setup — copy-paste configs for Cursor, Claude Desktop, Claude Code, VS Code, and Windsurf.
  • Agent Skills — discoverable skills published under /.well-known/ for agents to find and use the API on their own.

What is MCP? ​

The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools. APITube runs a hosted MCP server, so an assistant can translate a question like "find positive breaking news about Tesla in English" into a fully filtered News API request and return the articles — without you writing any code.

See MCP Server to connect in a minute.

AI-friendly data ​

Every article the News API returns is already enriched for LLM pipelines, so models get structured signal instead of raw text:

  • Sentiment — positive / negative / neutral scores at the title, body, and overall level.
  • Entities — people, organizations, locations, brands, and products extracted per article.
  • Categories & topics — IPTC categories and topic tags for precise filtering.
  • Readability — reading time and word/sentence counts.

These fields make the News API a strong source for RAG, summarization, and monitoring agents. Browse the sentiment and entity examples to see them in action.

Next steps ​