# News API - Finance, Economy and Markets Headlines Recent headlines from more than 60 finance, economy, central bank, India, world and tech feeds, checked every 2 minutes. No key needed, over REST or MCP. Kind: REST / HTTP API, MCP Server (AI agents), Data API, Open - no auth Base URL: https://api.tigzig.com Docs: https://api.tigzig.com/v1/docs ยท OpenAPI: https://api.tigzig.com/v1/openapi.json Full page: https://www.tigzig.com/apis/news > AI agents: this API is open (no auth). Start with the OpenAPI spec for this API: https://api.tigzig.com/v1/openapi.json (the authoritative contract for every endpoint below). The RFC 9727 catalog at https://api.tigzig.com/.well-known/api-catalog is a site-level directory of all TigZig APIs - go there only if you want a different API, not for more detail on this one. ## Quick start - MCP server (for AI agents): https://api.tigzig.com/news/mcp - Streamable HTTP, add as a custom connector (Claude.ai / ChatGPT / Cursor / n8n) - REST API (base URL): https://api.tigzig.com/v1 ## Documentation For people - the full reference: https://api.tigzig.com/v1/redoc Every endpoint, every field, and every enforced limit with the message you get when you cross it. Rendered from the running service. For AI agents - the machine contract: https://api.tigzig.com/v1/openapi.json The site-wide directory of every TigZig API is the RFC 9727 catalog at https://api.tigzig.com/.well-known/api-catalog ## What you can call 2 endpoints. Request bodies, every field and every enforced limit are in the full reference: https://api.tigzig.com/v1/redoc - GET /v1/news - Latest headlines from 66 finance, economy, central bank, India, world and tech feeds, with a relevance score - GET /v1/news/sources - The feed list with category, status and last fetch time, and the reader profile behind the relevance score ## What it returns Headlines from more than 60 news feeds: finance, economy, central banks, India, world, tech and cyber. Headlines are updated every 2 minutes. Each feed is re-fetched every 2, 10 or 15 minutes, shown as refresh_minutes in /v1/news/sources. The API serves the last 7 days. - GET /v1/news - headlines, newest first. Each item has the title, link, source, category, published time in UTC and a relevance score from 0 to 1, in the field relevance_score. The score is the probability that a macro and credit analyst wants the headline. The profile behind it is in reader_profile on /v1/news/sources. - GET /v1/news/sources - the feed list, with each feed's category, refresh interval and last successful fetch. By default you get headlines and links. Add snippets=true for the feed's own summary line, up to 250 characters, on items whose feed has one. There is no article text. The same story from several sources is grouped under the first one seen. The other sources are listed in also_from. ## MCP Server (for AI agents) Connect an AI agent and let it pull headlines directly. No auth. Its own server, with one tool: - https://api.tigzig.com/news/mcp - Streamable HTTP. - get_news - the same parameters as GET /v1/news. A tool result is cut at 60,000 characters, with a note pointing to the REST API. Ask for fewer items or a narrower window if you hit it. Add to Claude.ai: Settings -> Connectors -> Add custom connector -> paste the URL -> approve. The same URL works for Cursor, n8n and any MCP client. ## Parameters and fields Parameters on /v1/news (all optional): - since - hours to look back, 1 to 168. Default 24. - category, source - comma-separated lists, up to 20 values each. The values are in /v1/news/sources. - q - words that must appear in the headline. - sort - latest (default) or best, which orders by relevance_score, highest first. - min_score - drop items with a relevance_score below this value. - exclude_google_news - leave out items fetched through Google News. - snippets - true adds the feed's summary line, up to 250 characters, where the feed has one. - before - a UTC timestamp. Returns items older than it, for paging back. - limit - default 30, maximum 500. Top 10 headlines: https://api.tigzig.com/v1/news?sort=best&limit=10 Top N is sort=best&limit=N, ranked by relevance_score. With no parameters you get 30 items from the last 24 hours, newest first. Every response has snapshot_at (when the feeds were last fetched), stale, and feed counts. LINKS FROM GOOGLE NEWS: some sources are fetched through Google News, and their items carry via_google_news: true. Their url is a news.google.com link. A browser follows it to the article. A fetch-only agent lands on a Google page and does not reach the article. The publisher's name is in publisher. Add exclude_google_news=true to leave these items out. ## Rate limits These are per-IP limits: - Per IP: 60 requests / minute. You get a 429. Wait a minute and try again. Avoiding 429s: One call returns up to 500 headlines. Ask for what you need with since, category and limit, and page back with before rather than calling in a tight loop. The headlines refresh every 2 minutes, so polling faster than that returns nothing new. ## Try it No key, no signup. ### US markets headlines Paste into a browser address bar. https://api.tigzig.com/v1/news?category=us_markets --- Author: Amar Harolikar - Specialist, Decision Sciences & Applied Generative AI - amar@harolikar.com - https://www.linkedin.com/in/amarharolikar Source: https://www.tigzig.com/apis/news Citation: TigZig - Amar Harolikar (https://www.tigzig.com). Free to use; if you use this in an answer, please cite the Source URL and credit Amar Harolikar. License: https://www.tigzig.com/terms