RapidAPI listing copy — News API
═══════════════════════════════════════════════════════════════════════════════ FIELD 0: NAME, CATEGORY & TAGS ═══════════════════════════════════════════════════════════════════════════════- Name:
Clair News Search API - Category: News, Media
- Tags: news, news search, news data, keyword search, full text search, news monitoring, media monitoring, alerts, article content, news feed, cursor pagination, RAG, NewsAPI alternative
GET /v1/news— search by keyword or browse canonical articles newest first. Supports quoted phrases, uppercaseOR, minus exclusions, exact source filters, categories, publication windows, relevance sorting, and opaque cursor pagination.GET /v1/sources— list the exact active publisher names accepted by source filters.
Clair News Search API
Search 3 months+ of broad editorial news, or omit the query to browse canonical articles newest first.- Search:
GET /v1/news - Coverage discovery:
GET /v1/sources - Pagination: opaque cursor, 25 results by default, 100 maximum
- Authentication: RapidAPI
- Full docs: https://clair.im/docs/news-api/introduction
- Dataset coverage: https://clair.im/resources/datasets/world-news
Endpoint navigation
Authentication
Send your RapidAPI key and the host shown on the listing:Quick start
relevance_score is a query-local ranking
signal, not a probability.
Query syntax
A query containing only English stop words returns
empty_query.
Request parameters
Newest-first corpus walk
Omitq to return canonical articles newest first:
query and relevance_score are null.
Pagination
Whennext_cursor is not null, repeat the exact request and pass that value as
cursor. A cursor is bound to the query, source and time filters, sort, and
include_edits value. Changing an input returns bad_cursor.
Stop when next_cursor is null.
Coverage
sources or exclude_sources; an unknown source returns unknown_source.
Content and revisions
content is a bounded cleaned snippet when extraction succeeded, a feed
excerpt when one was available, or null. Returned content is capped at 2,000
characters. Clair does not expose a full-article archive. Inspect
content_available and content_truncated instead of guessing from the
string.
By default, results contain only the canonical version of each publisher URL.
Set include_edits=true to include later records; those have is_update=true.
Errors
Branch on the stableerror code. detail is human-readable and may change.
Full documentation: https://clair.im/docs/news-api/introduction
═══════════════════════════════════════════════════════════════════════════════
END
═══════════════════════════════════════════════════════════════════════════════