Continuously updated news from 234 countries and 60 languages, through one REST API or an MCP server. AI-categorized, deduplicated, and ready for your applications, your automations and your AI agents.
Free plan, no card required · 14-day trial on paid plans · Up and running in 5 minutes
A fully automated pipeline handles ingestion, AI enrichment, and delivery — so your team ships features instead of maintaining data pipelines.
Our crawler monitors 2,200+ news sources across every country and language, 24/7. New articles are detected within minutes of publication — no polling, no cron jobs, no scraper maintenance on your side.
Every article passes through our taxonomy pipeline: it receives a primary category, macro-category and language tag — normalized to a consistent schema so your queries work identically in every one of them.
Enriched articles are indexed and searchable within seconds. Each feed is read on its own rhythm, every 7 minutes when it is busy and at least every 45 minutes, around the clock, which brought in 67,540 articles on the last full day. Query by country, language, category, keyword, or date range — and get paginated, structured JSON back.
WorldFeedMonitor speaks the Model Context Protocol, the open standard AI assistants use to call tools. Connect it once and your agent searches and reads the news itself, in every country and language we cover, whenever a question needs it.
Nobody has to learn the filters. The agent looks up the codes, picks the countries and writes the keywords in the language the press uses.
Every article reaches the agent with its publisher and its link, and the server asks the agent to cite them, so a briefing points back to what was actually published.
It uses your existing API key. Each tool call counts as one request against your plan, and connecting or listing the tools costs nothing.
Included with every plan, Free included. Works with clients that let you set a request header: Claude Code, Cursor and VS Code, and Claude Desktop through mcp-remote. Browser connectors such as claude.ai and ChatGPT sign in with OAuth, which we do not offer yet.
Replace months of scraping, parsing, deduplication, and taxonomy work with a single integration that's already production-ready.
Stop maintaining country-specific crawlers. WorldFeedMonitor indexes 2,200+ news sources across 234 countries and 60 languages so you never have to build another scraper again.
Every article arrives pre-tagged with category, macro-category, and language. Build topic feeds, news alerts, or trend detection without training a single classification model.
Run full-text queries, boolean searches, and filtered lookups across the entire global news index — without provisioning, tuning or paying for a search cluster of your own.
Reliable infrastructure, usage monitoring, caching, and rate limits built in from the start — not retrofitted after launch.
Country, region, continent, language, category, publish date, author, images, and source domain — all consistent, all normalized, all ready to render directly in your UI.
Generate your API credentials, copy a request from the docs, and start querying immediately. No crawler setup, ingestion pipeline, or infrastructure provisioning required.
Regional news feeds, topic dashboards and content discovery, without integrating and maintaining 2,200+ RSS feeds yourself.
Follow topics, names and developments across countries and languages with keyword search, filtered by category, country, language and date.
Give MCP-compatible AI tools current news from around the world, without running a news collection pipeline of your own.
No surprise overages, no lock-in. Every paid plan starts with a 14-day free trial: cancel before it ends and you are not charged.
Build against the real API and see the shape of the data.
For teams shipping news-powered products to real users.
Every country we cover, at production volume.
For high-volume products and agents that need maximum throughput.
Can't find the answer here? Email us and we'll reply within 24 hours.