APITube

APITube turns the world's news into structured JSON for developers and AI agents. It aggregates articles from 500,000+ verified sources across 180+ countries in 60+ languages, and enriches every article before you see it: sentiment scores, extracted entities (people, organizations, locations, brands, events), IPTC categories, topics and industries. You filter instead of scraping and parsing HTML. Search runs through GET /v1/news/everything with 65+ filter parameters — keywords, language, country, source domain, source quality rank, sentiment range, entity IDs, media presence, date ranges — combined in one request. Other endpoints cover common shapes: /v1/news/top-headlines for breaking coverage, /v1/news/story for clustered articles, /v1/news/trends for trending values over time buckets, /v1/news/stream for a live Server-Sent Events feed, and /v1/suggest/entities to autocomplete entity IDs before filtering on them. Responses are JSON by default, with XML, CSV, TSV, XLSX, JSONL and Parquet exports on paid plans. Auth is one API key, passed as an api_key query parameter, an X-API-Key header or an Authorization: Bearer header. Webhooks and WebSocket dispatches push new matches to your endpoint as they get indexed. For AI assistants there is a hosted MCP server at https://mcp.apitube.io/ — no install, no local process, just the API key in the Authorization header. It exposes search_news and suggest as tools, so Claude, Cursor or VS Code can query live news directly. SDKs for Node.js, PHP and Python are on github.com/apitube; docs at https://docs.apitube.io. Typical uses: media and brand monitoring with sentiment attached, market and financial intelligence filtered by entity and industry, RAG pipelines that need a live news corpus, and products that ship a news feed without running a crawler.