Changelog

INFO

New features, changes and fixes in each APITube release.

1.0.55 — October 16, 2026

• [Cyber Security API] New Cyber Security API — a news feed of cyber crime and cyber warfare with vulnerability details on every article:
• Each article has a cyber block with its category (type), vulnerability IDs (cves), CISA KEV status (in_kev, exploited), affected vendors (vendors) and the highest CVSS score stated in the text (cvss). CVE IDs are read from the title, description or body. The CISA Known Exploited Vulnerabilities (KEV) catalog is refreshed every 24 hours. If it is unavailable, in_kev and exploited are null, not false.
  • Every filter, sort, facet, highlighting, fl and export option of /v1/news/everything works here and only narrows the cyber selection. fl=cyber.cves returns a single field; CSV, TSV and XLSX carry the block in a Cyber column, XML, JSONL and Parquet in cyber, RSS as item categories
  • 1 point per request that returns articles, the same plan limits as /v1/news/everything, available on every plan. API keys restricted to specific endpoints need the new cyber_security_news scope
  • New codes: ER0900 (400, invalid cyber.type) and the ER0901 warning (some stored articles were left out, the page may be short — keep following next_page). Start with the Quickstart

1.0.54 — September 11, 2026

• [Dashboard] Dataset exports — bulk export of articles to a file, straight from the dashboard (Datasets in the sidebar):
  • Choose filters from /v1/news/everything, set a published_at range of up to 366 days and select JSONL, CSV or Parquet. The page shows export progress. Download the completed file within 7 days. JSONL and CSV files use gzip compression; Parquet uses its own compression
  • Basic plan and above. Free and Starter see the page but cannot queue an export
  • Title searches (title, title_*, query) keep their 31-day window per request — the export splits a longer range into 31-day chunks automatically

1.0.53 — September 1, 2026

• [Plans] Pay-as-you-go now needs your explicit consent — a balance switch in the dashboard decides whether prepaid money can pay for requests:
  • The switch lives on Pay as you go and on Account → Plan & usage, and it is off by default. With it off, your balance is never touched: requests beyond the plan quota return 402 ER0176 instead of spending money, on REST, SSE, WebSocket and webhook deliveries alike
  • X-Balance-Remaining keeps reporting the real money on the account whether the switch is on or off; it says how much you have, not how much may be spent

1.0.52 — August 15, 2026

• [News API] Added the prompt parameter — describe the news you want in plain language and the API builds the filters for you, inside the same request:
• Send a sentence of 3–500 characters, such as prompt=Elon Musk and Tesla news for the last 10 days. The API converts it to filters before searching. Supported filters include:
  • Headline keywords and phrases; names in person.name, organization.name, location.name, brand.name, event.name, disease.name, disaster.name and sport.name
  • Categories, topics and industries: category.id, topic.id, industry.id
  • Publishers, authors, language and country: source.domain, author.name, language.code, source.country.code
  • Dates, tone and sorting: published_at.start, published_at.end, sentiment.overall.polarity, sort.by, sort.order
  • Yes-or-no filters: is_breaking, has_image, has_video, is_paywall, is_long_read; result count: per_page
  • Check meta.prompt for the original text, applied filters, ignored values with reasons, and whether the translation was cached. Unknown entity names are searched in headlines
• Parameters you send directly take priority. For example, language.code=de overrides a language chosen from the prompt. The unused value is reported with reason: explicit_param.
  • Accepted on /v1/news/everything, /v1/news/top-headlines, /v1/news/local, /v1/news/count, /v1/news/trends, /v1/news/stream, /v1/news/ws and the /v1/news/category · /topic · /industry · /entity endpoints. /v1/news/raw accepts it but only applies dates, sort.order and per_page; /v1/news/article, /v1/news/story and /v1/fact-check ignore it
  • 2 points per translation, charged only when that wording has not been parsed before — interpretations are cached for 24 hours, so a repeated prompt is free
  • A premium feature: Basic and above. On Free and Starter the request returns 403 ER0706 before the translation runs, so nothing is charged. Every filter a prompt would build stays available on every plan — prompt only writes them for you
  • New error codes: ER0800 (length outside 3–500), ER0801 (translation service unavailable, 502, nothing charged), ER0802 (nothing usable in the prompt); a plan without this feature returns ER0706 (403). Full walkthrough: Natural Language Search
• [Plans] Quota raised again on the same two tiers, prices unchanged: Basic 35,000 → 50,000 and Professional 90,000 → 150,000 requests per cycle. The annual plans (basic_yearly, professional_yearly) get the same allowance.

1.0.51 — August 9, 2026

• [News API] Free plan returns full links, images and descriptions. Article body text remains a preview:
  • href, image, description, source.domain, source.home_page_url, source.favicon, links[].url and media[].url are no longer truncated. Links on the Free plan are clickable, images load.
  • translations.en.title and translations.en.description are returned in full on every plan (they were cut to 30 and 200 characters on Free).
  • Entities, sentiment, categories, topics, industries and readability remain available in full on Free.
  • body and body_html remain a 200-character preview ending in ...[Upgrade subscription plan]. A paid plan with a live key returns the full text.
• [Plans] per_page is now capped per plan: 10 on Free, 50 on Starter, 250 on Basic and up. Requesting more returns 400 ER0171 with the plan ceiling in the message, and the default page size drops to the ceiling.
• [Plans] Quota raised on two tiers at the same price: Basic 20,000 → 35,000 and Professional 50,000 → 90,000 requests per cycle.

1.0.50 — August 5, 2026

• [Plans] New Starter plan — the entry paid tier, between Free and Basic:
  • $29/month or $279/year, renewing every 31 days (366 on the annual plan)
  • 10,000 requests per cycle, 3,000 streaming credits for SSE, WebSocket and webhook deliveries each (the annual plan carries 10,000 SSE deliveries)
  • 2 concurrent SSE and 2 concurrent WebSocket connections, 5 saved webhooks
• Paid plans include full article text with a live key, access beyond the first 5 pages, 50 requests per minute and /v1/fact-check. You can also choose to pay from your balance when plan credits run out.
• [Plans] Starter allows 50 articles per request and one third of Basic's streaming deliveries. It has the same connection and webhook limits as Basic. Higher plans allow larger page sizes.
• [Plans] The tier ladder is now Free → Starter → Basic → Professional → Corporate; upgrade and downgrade buttons in the dashboard follow that order.

1.0.49 — July 31, 2026

• [News API] Added English translations to every article — non-English coverage is now readable without a translation step of your own:
  • New translations block on the article object: translations.en.title and translations.en.description (the article body is not translated)
  • The block is returned always, with no request parameter to enable it; for English articles its fields are null, since the source text is already English
  • Original title and description are never replaced — the translation sits next to them, and language still reports the article's original language
  • English translations are available for 48 source languages
  • Available in JSON, JSONL/NDJSON, XML (<translations> node), CSV/TSV/XLSX (Translations column) and Parquet (translations column); request a single field with fl=translations.en.title
  • On the Free plan both translated fields are truncated: translations.en.description after 200 characters (exactly like description) and translations.en.title after 30 characters. The original title stays full on every plan

1.0.48 — July 29, 2026

• [News API] Breaking: title searches can cover at most a 31-day published_at window per request:
  • Applies to title, title_starts_with, title_ends_with, title_pattern and the boolean query parameter, on every plan and every endpoint, including exports, SSE stream and WebSocket
  • A title search with published_at.start / published_at.end more than 31 days apart now returns 400 ER0110. The same applies to an open-ended published_at.start older than 31 days, since the window then runs to the current time
  • A title search without any published_at dates keeps working and searches the last 31 days. The response says so explicitly: warning ER0366 in meta.warnings, plus an x-query-window-clamped: 31 header for export formats that carry no JSON body
  • The single-day published_at parameter is unaffected, and requests that do not match on the title (category.id, entity.id, source.domain, language.code and the rest) keep their unlimited range
  • Previously this window was narrowed silently and only on the Free plan; it is now an explicit, uniform limit. To search a longer period, issue one request per month-sized window

1.0.47 — July 10, 2026

• [MCP Server] MCP Server update — revised tooling for AI agents and MCP clients (Claude Desktop, Cursor, and others):
  • The news-search tool is now search_news (renamed from everything) — a clearer name for agents picking tools autonomously; filters and behavior are unchanged
• suggest finds the ID for a name or the start of a name. Use the returned ID in search_news: numeric entity.id or industry.id, an IPTC category.id such as medtop:…, or a text topic.id. Agents can look up these IDs before searching.
  • Four ready-made prompts for common workflows: monitor_company (coverage digest for a company), topic_sentiment (how a topic is covered), breaking_news (what is breaking right now), and compare_coverage (side-by-side coverage of two subjects)
  • API-key restrictions are now enforced at the MCP layer too: per-endpoint scopes, key expiration, and IP allowlists are checked on every tool call

1.0.46 — July 3, 2026

• [News API] Added Fact Check (/v1/fact-check) to check factual claims against recent news:
  • Check a single statement (claim), a full article's text (text), or any article already in the archive by its id — claims for text/id are extracted automatically
  • Every claim returns a verdict on an 8-level scale (true, mostly_true, mixed, misleading, mostly_false, false, unverified, outdated), a confidence score, a short explanation, and the supporting/refuting evidence articles
• Set max_claims to limit how many claims are checked and evidence_per_claim to limit how many articles support each check. Both default to 5, and values above the allowed maximum are reduced.
  • Paid feature: each request costs 5 points and allows up to 10 requests per minute. It is unavailable on Free
  • Both GET and POST supported

1.0.45 — June 26, 2026

• [News API] Added Entity-Level Sentiment — sentiment measured toward each detected entity, not only the article as a whole:
  • Every entity in the entities array now carries a sentiment block: score (−1 to 1), polarity (positive, negative, neutral), and per-mention counts mentions (positive, neutral, negative)
  • New filters to narrow results by how an entity is portrayed: entity.sentiment.polarity (positive / negative / neutral), entity.sentiment.score.min, entity.sentiment.score.max (range −1 to 1)
  • Combine with entity.id or a name filter (person.name, organization.name, …) to find positive or negative coverage of a specific entity. For example, find articles with negative coverage of one company. Use without an entity filter to match any entity with that sentiment.
  • New error codes: ER0290 (invalid entity.sentiment.polarity), ER0291 (invalid entity.sentiment.score.min), ER0292 (invalid entity.sentiment.score.max)

1.0.44 — June 19, 2026

• [News API] Added Webhooks — push delivery of new articles to your own endpoint as they are published, instead of polling:
  • Each webhook is a saved query: a destination url plus the same filters as /v1/news/everything (language, category, source, sentiment, entities, …). When a newly indexed article matches, APITube POSTs it to your URL.
  • Signed deliveries — every request carries X-Webhook-Signature (HMAC-SHA256 over timestamp + "." + body), plus X-Webhook-Id and X-Webhook-Timestamp, so you can verify authenticity.
  • Automatic retries with exponential backoff (1m → 5m → 30m → 2h → 12h, up to 5 attempts); a subscription that keeps failing is auto-disabled.
  • Create webhooks and inspect delivery logs (status, attempts, response code) in the dashboard. Each delivered article costs 1 point.
  • Per-plan limits: Free 1 · Basic 5 · Professional 20 · Corporate 100 webhooks.

1.0.43 — June 12, 2026

• [News API] Added Local News endpoint (/v1/news/local) — find news near a point on the map, with the closest results first:
  • Set the search centre with lat and lng, or enter a name in place. Add country if the place name exists in more than one country.
  • radius in kilometers (default 50, up to 20000); each article includes distance_km and a nearest_location block (matched location entity, coordinates, country)
  • sort — distance (default), published_at, or relevance
  • Relevance ranking combines distance, publisher rank (Open PageRank), article age and importance. Choose a ranking preset (balanced, proximity, authority, fresh) or adjust the factors with w.distance, w.opr, w.recency and w.important.
  • Supports all /v1/news/everything filters except location.*, plus fl field selection, pagination and every export format.
• Optional insights summaries describe coverage in the search area. Choose from mood, hotspots, events, entities, bias, sources, timeline, top_categories, top_topics, breaking, velocity and movers.
  • Both GET and POST supported; billing is 1 point for the listing plus 1 point per requested insight block

1.0.42 — June 5, 2026

• [News API] Added Reference Endpoints — searchable directories of the entities behind the news, each with a coverage block and the latest articles:
  • People (/v1/people, /v1/people/{id}) — public figure profiles with background details (title, bio, nationality, social profiles)
  • Companies (/v1/companies, /v1/companies/{id}) — organizations and brands (headquarters, founded, website, key people, subsidiaries)
  • Journalists (/v1/journalists, /v1/journalists/{id}) — profiles group records with the same author name and list where they publish in outlets
  • Coverage on each profile: article counts, monthly timeline, sentiment breakdown, momentum (last 30 days vs. previous), top sources / topics / countries / languages, related (co-mentioned) entities, and recent_articles
  • Each request costs 1 point, charged only when the response contains at least one result

1.0.41 — May 29, 2026

• [News API] Added Raw Articles (/v1/news/raw) for the earliest available articles in a basic format. Articles remain available for roughly 24 hours:
  • Article text in body and original HTML in body_html, just as on /v1/news/everything
• Publisher details in source: domain, type, bias, Open PageRank, location and favicon. Original article keywords appear in keywords.
  • Filter by source.id / ignore.source.id or publication dates, and sort with sort.by / sort.order
  • Both GET and POST supported; each request costs 1 point (charged only when the response contains at least one article)

1.0.40 — May 22, 2026

• [News API] Added Suggest Endpoints — suggest filter values as a user types. Each endpoint accepts prefix and returns matching IDs for its filter:
  • Category Suggestions (/v1/suggest/categories) — id maps to the category.id filter
  • Topic Suggestions (/v1/suggest/topics) — id maps to the topic.id filter
  • Industry Suggestions (/v1/suggest/industries) — id maps to the industry.id filter
  • Authentication via X-API-Key header or api_key query parameter

1.0.39 — May 15, 2026

• [News API] Added Count Articles endpoint (/v1/news/count) — returns only the number of articles matching your filters, without loading any article content. Accepts the same filters as /v1/news/everything; ideal for dashboards, sample-size estimation, and progress bars.

1.0.38 — April 17, 2026

• [News API] Added Node.js SDK — official Node.js client library for APITube News API:
  • Call all News API endpoints from your application
  • Install via npm: npm install @apitube/news-api
  • Full support for all query parameters, filtering, and pagination
  • Promise-based HTTP client with TypeScript definitions

1.0.37 — April 10, 2026

• [News API] Added PHP SDK — official PHP client library for APITube News API:
  • Call all News API endpoints from your application
  • Install via Composer: composer require apitube/news-api-php
  • Full support for all query parameters, filtering, and pagination
  • PSR-compatible HTTP client

1.0.36 — March 30, 2026

• [News API] Added n8n Integration Guide — connect APITube News API to n8n with step-by-step instructions:
  • Setting up APITube credentials in n8n (Header Auth with X-API-Key)
  • Ready-to-import workflow: News Monitoring — scheduled polling with keyword, language, and sentiment filters
  • Ready-to-import workflow: News to Slack / Telegram — automatic alerts with formatted article cards
  • Import the AI News Digest workflow to summarise each day's news with OpenAI and email the result
  • Common parameters and tips for pagination, rate limits and avoiding duplicate articles

1.0.35 — March 18, 2026

• [News API] Added API Reference — interactive API documentation with request builder, response examples, and schema explorer
• [News API] Added OpenAPI Specification — an OpenAPI 3.0 file for generating client libraries, importing requests into Postman and using other API tools

1.0.34 — March 5, 2026

• [News API] Added SSE Real-Time Stream — new endpoint /v1/news/stream for receiving articles in real time via Server-Sent Events:
  • Live article delivery as they are published — no polling required
  • All /v1/news/everything filters are supported (language.code, category, source.domain, sentiment.*, etc.)
  • Field selection via fl parameter (e.g., fl=id,title,source.domain)
  • Automatic reconnection support via Last-Event-ID header — no articles missed during disconnects
  • SSE events: article (new article with JSON data), error (server error with code and message), heartbeat comments every 30 seconds
  • Authentication via X-API-Key header or api_key query parameter
  • Billing: 1 credit per delivered article
  • Compatible with native EventSource API in browsers, curl, Python, and any SSE client

1.0.33 — March 1, 2026

• [News API] Added API Key Restrictions — choose which IP addresses and HTTP Referrers can use each key:
  • Configure allowed IP addresses (exact match or wildcard patterns like 10.0.0.*)
  • Configure allowed HTTP Referrers (exact match or wildcard patterns like *.example.com)
  • An empty list leaves access unrestricted
  • New error codes: ER0601 (IP not allowed), ER0602 (domain not allowed)
  • Restrictions are validated before rate limiting — blocked requests do not consume quota

1.0.32 — February 19, 2026

• [News API] Added article export formats: Parquet file, JSONL.

1.0.31 — February 5, 2026

• [News API] Added Social Shares - new object in article response with social media sharing metrics:
  • shares.total - total number of shares across all platforms
  • shares.facebook - number of Facebook shares
  • shares.twitter - number of Twitter/X shares
  • shares.reddit - number of Reddit shares
• [News API] Added Social Shares Sorting - new sorting parameters:
  • shares.facebook.min / shares.facebook.max - sort by Facebook shares
  • shares.twitter.min / shares.twitter.max - sort by Twitter shares
  • shares.reddit.min / shares.reddit.max - sort by Reddit shares

Note: Share counts update during the first 2 hours after publication. A newly published article may show a low count; check again later for the full value.

1.0.30 — January 22, 2026

• [News API] Added relative dates (Solr-style Date Math) for all date and time parameters:
  • NOW keyword: NOW returns current time, NOW+1DAY, NOW-1WEEK for offsets
  • Truncation/Rounding: NOW/DAY (start of today), NOW/MONTH (start of month), NOW/YEAR (start of year)
  • Combined: NOW-1DAY/DAY (start of yesterday), NOW+1MONTH/MONTH (start of next month)
  • Short aliases: 1w (1 week ago), 2m (2 months ago), 3d, 4h, 5y
  • Direction modifiers: +1w (future), -2m (past, explicit)
  • Full units: 1WEEK, 2MONTHS, 3DAYS, 24HOURS, 1YEAR
  • Works with published_at.start, published_at.end, compare_window, and facet.range.gap
  • Examples: published_at.start=NOW-7DAYS, published_at.start=NOW/WEEK, compare_window=1w

1.0.29 — January 8, 2026

• [News API] Added new endpoint /v1/news/trends - get trending values for a specific field, ranked by article count. Discover the most popular entities, categories, topics, industries, sources, countries, languages, and authors across your filtered result set.
  • field - the field to group and count articles by. Allowed fields: source.id, category.id, topic.id, industry.id, entity.id
  • limit - number of top results to return (default: 10, max: 100)
  • Results include names and details for entities, categories, topics, industries, sources, countries, languages and authors

1.0.28 — December 25, 2025

• [News API] Added Geo-location Search - filter articles by location:
  • Circle Search - filter by radius around coordinates:
    • location.lat - latitude coordinate (e.g., 52.52 for Berlin)
    • location.lng - longitude coordinate (e.g., 13.40 for Berlin)
    • location.radius - search radius in kilometers (e.g., 50 for 50km radius)
    • location.radius.min - minimum radius for ring search (exclude inner circle)
    • location.radius.max - maximum radius for ring search
  • Bounding Box Search - filter within rectangular areas:
    • location.bbox - format: minLat,maxLat,minLng,maxLng (e.g., 40.0,45.0,-74.0,-73.0 for NYC area)
  • Articles with coordinates - filter by whether coordinates are available:
  • has_location_geo=1 - only articles with coordinates
  • has_location_geo=0 - only articles without coordinates
  • Example: /v1/news?location.lat=52.52&location.lng=13.40&location.radius=50
  • Use it for local news feeds and regional event monitoring

1.0.27 — December 11, 2025

• [News API] Added Media Richness Sorting - new sort.by=media_richness parameter to sort articles by total media content:
  • Ranks articles by their combined image and video count
  • Use it to find articles with images and videos
  • Use with sort.order=desc to get articles with the most images and videos first
• [News API] Added reading-time filters to select short, medium-length or long articles:
  • is_quick_read=1 - articles with read time ≤ 2 minutes (news briefs, updates)
  • is_medium_read=1 - articles with read time 3-7 minutes (standard articles)
  • is_deep_dive=1 - articles with read time ≥ 10 minutes (in-depth analysis, long-form journalism)
• [News API] Added High Quality Content Filter - use is_high_quality to find articles that meet source, image and author requirements:
  • is_high_quality=1 - articles that are not duplicates, have a source rank ≥ 5, include images and have an identified author
  • Use it to select articles for your feed

1.0.26 — November 27, 2025

• [News API] Added Clickbait Detection Filter - use is_clickbait to find headlines whose tone differs sharply from the article body:
  • is_clickbait=1 - find articles where title polarity differs from body AND title has strong sentiment (|score| > 0.5)
  • is_clickbait=0 - find articles with a similar tone in the headline and body
  • Use it to find attention-grabbing headlines whose tone differs from the article
• [News API] Added Sentiment Gap Filter - measure the difference between title and body sentiment scores:
  • sentiment_gap.min - minimum gap between title and body sentiment (0-2 range)
  • sentiment_gap.max - maximum gap between title and body sentiment (0-2 range)
  • Use it to compare the tone of headlines and article text

1.0.25 — November 13, 2025

• [News API] Added Readability Filters - filter articles by reading difficulty and target audience:
  • Flesch-Kincaid Grade: readability.fk_grade, readability.fk_grade.min, readability.fk_grade.max - filter by US grade level (0-30)
  • Flesch Reading Ease: readability.ease, readability.ease.min, readability.ease.max - filter by ease score (0-100, higher = easier)
  • Automated Readability Index: readability.ari, readability.ari.min, readability.ari.max - filter by ARI score
  • Difficulty Level: readability.difficulty - filter by beginner, intermediate, advanced, expert
  • Target Audience: readability.audience - filter by children, general, professional, academic
  • Reading Age: readability.age, readability.age.min, readability.age.max - filter by recommended reader age (6-22)
  • Convenience filters: is_easy_read (FRE ≥ 60), is_difficult_read (FRE < 40)
• [News API] New readability object in article response with all readability metrics

1.0.24 — October 30, 2025

• [News API] Added 12 image filters for size, orientation and other image properties:
  • Orientation filters: is_landscape_media, is_portrait_media - filter by image orientation
  • Resolution filters: has_fullhd_images (1920px+), has_4k_images (3840px+), has_mobile_optimized_images (320-800px)
  • Social media ready: is_instagram_ready (1080px+, square/4:5 ratio), is_twitter_card_ready (800px+, landscape)
  • Content filters: has_multiple_images (2+ images), has_thumbnail (small images ≤300px), has_social_share_image (Open Graph compatible: 1200x630+)
  • Quality filters: has_consistent_image_sizes (uniform image sizes, variance <200px)
  • Mixed content: has_mixed_media (articles with both images and videos)

1.0.23 — October 16, 2025

• [News API] Added 5 new facet fields:
  • content.length - categorize articles by length (short, medium, long)
  • published.weekday - distinguish weekday vs weekend articles
  • published.time_of_day - group by time of day (morning, afternoon, evening, night)
  • sentiment.strength - classify sentiment intensity (strong, moderate, neutral)
  • entity.id - count articles by entity

1.0.22 — October 2, 2025

• [News API] Added Range Facets to count articles within date or numeric intervals. Use the counts for publication timelines, sentiment charts and reading-time charts.
• [News API] Range Facet parameters:
  • facet.range - Enable/disable range faceting
  • facet.range.field - Field to create ranges on (published_at, sentiment.overall.score, read_time, etc.)
  • facet.range.start - Start value for range
  • facet.range.end - End value for range
  • facet.range.gap - Gap/interval between ranges (e.g., +1DAY, +1WEEK, 0.1 for numeric)
• [News API] 8 range facet fields now available:
  • Date fields: published_at - analyze article distribution over time
  • Sentiment scores: sentiment.overall.score, sentiment.title.score, sentiment.body.score
  • Content metrics: read_time, media.images.count, media.videos.count
  • Source quality: source.rank.opr

1.0.21 — September 18, 2025

• [News API] Added Faceting support - count articles grouped by field. Use the counts in filters and dashboard charts.
• [News API] 24 facet fields now available including:
  • Source fields: source.id, source.country.id, source.bias, source.rank.opr
  • Classification: category.id, topic.id, industry.id
  • Language & Author: language.id, author.id
  • Sentiment: sentiment.overall.polarity, sentiment.title.polarity, sentiment.body.polarity
  • Article attributes: is_duplicate, is_free, is_important
  • Media content: media.images.count, media.videos.count
  • Reading time: read_time
  • Publication dates and times: published.year, published.month, published.day_of_week, published.hour

1.0.20 — September 4, 2025

• [News API] Added Highlighting support - get text snippets with matching search terms highlighted. Use these snippets on search result pages.
• [News API] Highlight related words — highlighting now includes synonyms and word forms. For example, a search for "innovation" can highlight "innovative" and "innovator".
• [News API] Highlighting parameters:
  • hl - Enable/disable highlighting
  • hl.fl - Fields to highlight (title, description, body)
  • hl.fragsize - Snippet size (50-500 characters)
  • hl.snippets - Number of snippets per field (1-10)
  • hl.tag.pre / hl.tag.post - Custom highlight tags (default: <em> / </em>)

1.0.19 — August 21, 2025

• [News API] Added fl (field list) to choose which fields a response returns. Use dots to select nested fields, for example fl=id,title,source.name,sentiment.overall.score.

1.0.18 — August 7, 2025

• [News API] Added new endpoint /v1/suggest/entities - find entities as a user starts typing a name. Returns matches with their details, types and Wikipedia/Wikidata links.

1.0.17 — July 24, 2025

• [News API] Added new endpoint /v1/balance - check your API key balance and remaining credits. Returns api_key, points, and plan information.
• [News API] Added rate limiting to /v1/balance endpoint - limited to 30 requests per minute per API key to prevent abuse.

1.0.16 — July 10, 2025

• [News API] Added disaster.name and ignore.disaster.name to include or exclude articles about named disasters.
• [News API] Added disease.name and ignore.disease.name to include or exclude articles about named diseases.
• [News API] Added event.name and ignore.event.name to include or exclude articles about named events.

1.0.15 — June 26, 2025

• [News API] Added body_html, the article text with HTML formatting. Available since December 30, 2025.

1.0.14 — June 12, 2025

• [News API] Added is_breaking to articles: true for breaking news, false otherwise.
• [News API] Added read_time to articles: estimated reading time in minutes.

1.0.13 — May 29, 2025

• [News API] Added source.bias to articles: the publisher's bias, left, right or center.
• [News API] Added List of Entities. Use the listed entities to filter articles.

1.0.12 — May 15, 2025

• [News API] Updated categories. See the full list of categories in the documentation.

1.0.11 — May 1, 2025

• [News API] Added new topics with the latest news from the commodities market.

1.0.10 — April 17, 2025

• [News API] Fixed rate limiting across all plans.

1.0.9 — April 3, 2025

• [News API] Added is_paywall: true for articles behind a paywall, false otherwise.
• [News API] Filter articles by paywall status. Example: https://api.apitube.io/v1/news/everything?is_paywall=0. More examples in the documentation.
• [News API] Filter by duplicate articles. Example: https://api.apitube.io/v1/news/everything?is_duplicate=1. More examples in the documentation.
• [News API] Added favicon to source: the URL of the publisher's favicon.

1.0.8 — March 20, 2025

• [News API] Get news by topics: crypto news, movie news, video game news

1.0.7 — March 6, 2025

• [News API] Added article export formats: XML, XLSX and CSV.

1.0.6 — February 20, 2025

• [News API] Added story_id to articles as the story identifier.

1.0.5 — February 6, 2025

• [News API] Added links and videos to articles. links lists related article URLs.

1.0.4 — January 23, 2025

• [News API] Added the opinion resource type for articles that express the writer's opinion rather than report news.

1.0.3 — January 9, 2025

• [News API] Added the search filter to find articles by matching their title or content. Example: https://api.apitube.io/v1/news/everything?search=bitcoin.
• [News API] Raised the page-size limit for paid users to 1,000 articles.

1.0.2 — December 26, 2024

• [News API] Added new integrations: Dart, Elixir, Haskell, Lua, Scala, Rust.

1.0.1 — December 12, 2024

• [News API] Added new categories.

• [News API] Article body HTML is now trimmed.