Decrypt Crypto News Finder by Category & Keyword

Decrypt publishes crypto, Web3, AI, policy and market news across a deep archive and five languages. This actor searches that archive by keyword, browses 47 categories, tags, 121 coins or authors, looks up specific articles by ID or URL, and returns full article records with HTML and plain text, TL;DR bullets, editors, author profiles, images, taxonomy and reading metrics. No account, cookies or proxy setup is required.

What this actor does

• Seven modes: keyword search, browse by category, lookup by ID/URL, latest articles, browse by tag, browse by coin and browse by author
• Deep archive access with server-side date ranges, newest/oldest/relevance sorting and up to 10,000 articles per run
• Full article data: HTML body, plain-text body, TL;DR, editors, author bios and social profiles, tags, coins, collections, distinctions and images
• Five locales: English, Spanish, French, Arabic and Simplified Chinese, with the correct localized article URL
• Batch lookups: paste article IDs or decrypt.co URLs, including localized and /resources/ links
• Scheduled monitoring: onlyNew skips articles returned by earlier runs
• Empty fields are omitted from every record instead of being returned as null

Output per article

• id — Decrypt article ID
• title, slug, excerpt, blurb — headline and summaries
• locale, status, publishedAt, modifiedAt — language, publication status and ISO timestamps
• content — full article body as HTML
• contentText — plain-text article body for search, NLP and summaries
• wordCount, readingTimeMinutes — body word count and reading time in minutes
• readingTime — original reading-time label from the source
• tldrTitle, tldrContent, tldrText — "In brief" heading, HTML bullets and plain-text bullets
• authors[], authorSlugs[] — byline names and slugs
• authorDetails[] — name, slug, title, bio, location, twitter, linkedin, avatarUrl, roles
• editors[], editorSlugs[] — editor names and slugs
• category, categorySlug, categoryParent, categoryParentSlug — leaf category and its parent section
• tags[], tagSlugs[] — topic names and filterable slugs
• coins[], coinSlugs[], coinSymbols[] — mentioned coins, slugs and ticker symbols
• collections[], collectionSlugs[] — editorial collections
• distinctions[], distinctionSlugs[] — editorial distinctions
• celebrities[], celebritySlugs[] — people linked to the article
• featuredImageUrl, featuredImageAlt, featuredImageWidth, featuredImageHeight — featured image
• featuredVideoUrl — featured video URL when the article has one
• metaDescription, socialImageUrl — SEO description and social-sharing image
• sourceUrl, canonicalUrl — localized article URL
• recordType — always "article"
• scrapedAt — timestamp of when the record was scraped

Input

Field	Type	Default	Description
mode	string (enum, required)	search	search, browseByCategory, byId, trending, byTag, byCoin or byAuthor
searchQuery	string	""	Keyword matched against titles, excerpts and bodies (mode search)
category	string (enum)	""	One of 47 category slugs; empty means any category
categories	array	[]	Several category slugs at once
tagSlugs	array	[]	Tag slugs, e.g. bitcoin, regulation
coinSlugs	array	[]	Coin slugs, names or tickers, e.g. bitcoin, ethereum, SOL
authorSlugs	array	[]	Author slugs, e.g. jose
articleIds	array	[]	IDs or decrypt.co URLs (mode byId)
locale	string (enum)	""	en, es, fr, ar, zh-hans; empty means all languages
dateRangeFrom	string	–	Only articles published on or after this date (YYYY-MM-DD)
dateRangeTo	string	–	Only articles published on or before this date (YYYY-MM-DD)
sortBy	string (enum)	newest	newest, oldest or relevance
includeContent	boolean	true	Include HTML/plain-text bodies, TL;DR content and word count
onlyNew	boolean	false	Skip article IDs returned by earlier runs
maxItems	integer	10	Maximum articles to return (1–10,000)
proxyConfiguration	object	off	Optional override; the actor works without a proxy

Example: keyword search with relevance and a date window

{
  "mode": "search",
  "searchQuery": "etf inflows",
  "sortBy": "relevance",
  "dateRangeFrom": "2026-01-01",
  "dateRangeTo": "2026-03-31",
  "locale": "en",
  "maxItems": 200
}

Example: browse a category in Spanish without article bodies

{
  "mode": "browseByCategory",
  "category": "markets",
  "locale": "es",
  "includeContent": false,
  "maxItems": 100
}

Example: look up a mixed batch of IDs and URLs

{
  "mode": "byId",
  "articleIds": [
    "382316",
    "https://decrypt.co/373492/bitcoin-64000-inflacion-ipc-junio-fed-tasas-interes",
    "https://decrypt.co/resources/what-is-eip-7702-the-ethereum-upgrade-behind-the-6-4b-binance-hack"
  ]
}

Example: monitor one coin for new stories

{
  "mode": "byCoin",
  "coinSlugs": ["solana"],
  "dateRangeFrom": "2026-10-01",
  "onlyNew": true,
  "sortBy": "newest",
  "maxItems": 50
}

Use cases

• Crypto news monitoring — track coverage of a coin, project, company or person across the archive
• NLP and sentiment research — feed clean contentText and tldrText into your own models
• Media intelligence — compare category, tag and author coverage over specific date windows
• Content aggregation — power a newsletter, digest or research dashboard with full articles
• Editorial archiving — build a localized archive across English, Spanish, French, Arabic and Chinese
• Scheduled alerts — use onlyNew to pick up only stories published since the last run

FAQ

Is this affiliated with Decrypt? No. This is an independent, third-party actor that reads publicly available Decrypt article pages and feeds. It is not operated by or affiliated with Decrypt.

Do I need an account, API key or cookie? No. The actor runs without credentials and without proxy configuration.

Does it return full article text or just a summary? Both. content is the complete HTML body and contentText is a plain-text version. excerpt, blurb and the TL;DR fields are also included when the article has them.

Why are some fields missing from a record? Not every article has a coin, collection, distinction, editor, TL;DR or video. Fields with no value are omitted rather than returned empty or null. Decrypt's coin filter can also match stories whose article record has no explicit coin relation, so a byCoin run may include rows without coins or coinSlugs; the actor reports the article honestly instead of inventing a coin.

What do tagSlugs, coinSlugs and authorSlugs expect? Tag and author filters use the slugs returned in each record, not display names. Coin filters accept slugs plus common names and ticker symbols, which are mapped to slugs automatically.

How fresh is the data, and is trending a real trend signal? The archive is read live at run time. trending is the legacy name for the newest-first mode; it returns the latest published articles, not a proprietary popularity score.

How many articles can one run return? Up to 10,000 with maxItems. The source exposes a very large archive; for very broad queries, split the work by date range or category to keep each run focused.

Do localized articles have working URLs? Yes. sourceUrl and canonicalUrl use the article's own locale path, including /es/, /fr/, /ar/ and /zh-hans/.

Are sentiment scores or AI analysis included? No. Those values are not published by the source, and the actor does not fabricate them. Use contentText or tldrText with your own analysis pipeline.

Can I run this on a schedule? Yes. Set onlyNew to true and the actor will skip article IDs returned by earlier runs, so each scheduled run returns only unseen stories.