Universal Search APIFan out one query across social platforms and research sources with one SocialCrawl API key. Ranked, fused results in a single JSON envelope. Reads are priced in credits.
3 active meta endpoints. Everywhere, forums, and news.
Three peer meta endpoints under /v1/search. Cross-platform everywhere search, a forums lane, and planned multi-country news. Social and forum sources first. Not a generic open-web SERP alone.
/v1/search/everywhereOne query fans out across social and research sources in parallel. Planner, fusion, rerank, and clusters. Optional top comments on enrichable posts. JSON or SSE. Flat 20 credits per call.
query, lookback_days, from_date, to_date, sources, exclude
/v1/search/forumsFused threads from Reddit, Hacker News, and Korean forums (지식iN, 카페). Hero threads carry top comments by default. Flat 10 credits with coverage-based partial refund. Sync JSON.
query, sources, exclude, comments, timeframe, lookback_days
/v1/search/newsPlans angles, localizes per country, then fans google_news legs (up to 12). Deduped NewsArticle rows with per-leg provenance. JSON or SSE. Base 2 credits plus 1 per leg that returns articles.
query, countries, time_range, from, to, publisher, depth, max_legs
Returns one ranked, clustered result set for a query run in parallel across Reddit, X, YouTube, TikTok, Instagram and more, with top comments attached.
Use it when you want a topic's coverage everywhere at once instead of calling each platform's search endpoint yourself.
query · Search query (1–512 chars)
$ curl https://www.socialcrawl.dev/v1/search/everywhere?query=kanye+west&lookback_days=30&sources=reddit%2Cyoutube%2Cgithub \
-H "x-api-key: sc_YOUR_API_KEY"// Running this live needs your own API key — hit "Try it" to see an example responseUniversal Search is a SocialCrawl meta surface. You call GET /v1/search/… with an API key, pay the flat or metered cost on live work, and get one envelope. No per-platform OAuth for the fan-out.
Send your key in the x-api-key header. The same key works for Reddit, TikTok, Google Trends, Tavily, and every other catalog route.
All three routes are GET. Pass query plus lane options (lookback, sources, countries, timeframe). We validate before charging. Unknown source names return 400.
Everywhere is flat 20. Forums is flat 10 with partial refund when coverage fails. News is metered 2-14 (plan base plus successful legs). Cache hits cost 0 where applicable. Hard fails refund.
Every response is success, data, credits_used, credits_remaining, request_id, cached. Everywhere and news also support Accept: text/event-stream for progressive chunks.
Most products start with everywhere for breadth, drop into forums when thread quality matters, then news when press coverage is the signal.
GET /v1/search/everywhere?query=…Fused items[], clusters, per-source rollup, optional top_comments
One call replaces a dozen platform searches and client-side merge.
GET /v1/search/forums?query=…Fused forum items[], raw buckets, clusters, question_share style metrics
VoC and support language live in long threads, not short posts alone.
GET /v1/search/news?query=…&countries=…plan, legs with query_source, deduped NewsArticle items[]
Multi-country Google News legs without building the planner yourself.
GET /v1/search/everywhere
?query=social+media+api
&lookback_days=30
Host: www.socialcrawl.dev
x-api-key: sc_your_api_key_here
Accept: application/json
# SSE stream
# Accept: text/event-stream{
"success": true,
"data": {
"items": [],
"clusters": [],
"sources": {}
},
"credits_used": 20,
"credits_remaining": 9999,
"request_id": "req_…",
"cached": false
}Each lane keeps a stable meta envelope. Field names stay consistent so agents and dashboards can share parsers across the three routes.
items[] fused and ranked, clusters[], sources / items_by_source, top_comments on enrichable posts
fused items[], raw per-source buckets, thread clusters, computed block (question_share, top_communities)
plan + legs[] provenance, deduped NewsArticle items[] with country and angle context
Same gateway as /v1 platform routes. Meta fetchers run the planner, fan-out, fusion, and billing settlement inside SocialCrawl.
Next.js catch-all routes into the Hono social API. We mint a request_id, authenticate the key, then enforce rate limit and concurrency for the key.
Registry lookup finds search/everywhere, search/forums, or search/news. Params are checked first. Invalid input returns 400 with no charge. Flat lanes debit up front. News holds a ceiling and settles to actual legs.
Everywhere and forums run multi-source fan-out with fusion and clustering. News plans angles, localizes per country, and calls google_news legs. SSE streams intermediate chunks when Accept is text/event-stream (everywhere and news).
Results land in the meta Analytics envelope, credits settle (including news refunds for unused ceiling and empty legs), and the success envelope is logged for billing audit.
Billing rules that matter in production
Universal Search fans out across social and research sources, not the open web alone. That is the product difference versus Exa, Tavily, or Firecrawl used by themselves.
One-query social research across platforms, forums, and multi-country news. Built for agents, dashboards, and VoC jobs that need fused social signal, not a single SERP page.
Internal meta fetchers call registered platform endpoints in parallel. Everywhere covers social and research sources. Forums targets Reddit, Hacker News, and Korean forums. News plans google_news legs per country and angle.
A unified JSON envelope with fused items, clusters or legs, and credit fields. Everywhere and news can stream over SSE. Forums returns sync JSON with inline comments on hero threads.
Not a replacement for full web crawl suites. Not per-platform write access. Forums is sync-only in v1 (no SSE yet). News is Google News editions, not arbitrary HTML scrapes.
The jobs this API is most often used for.
Cross-platform topic sweeps and multi-country news
Callers use everywhere for breadth, forums when long-form discussion matters, and news when press editions across countries matter. SSE shows up on long everywhere and news runs. Social-first fan-out with flat 20 and 10 credit lanes on everywhere and forums, plus metered multi-country news. Open-web tools alone do not own this social spine.
Everywhere is a multi-source pipeline; expect longer latency than a single platform read, often many seconds with progressive SSE. Forums is sync. News scales with leg count.
The jobs Universal Search data is most often used for. Each one is a full recipe with the endpoint chain and pricing.
Common ways teams put this data to work, and the stack each one tends to run.
everywhere for brand and product language across platforms, then forums when thread depth matters.
search/news with country CSVs for localized coverage. Stream plan_refined then legs as they settle.
One meta call instead of N platform tools. Same SocialCrawl key for deepen routes after fusion.
Everywhere is flat 20 credits. Forums is flat 10. News is metered from 2 to 14 credits. Set Accept to text/event-stream for SSE on everywhere and news.
curl "https://www.socialcrawl.dev/v1/search/everywhere?query=social+media+api&lookback_days=30" \
-H "x-api-key: sc_your_api_key_here"curl "https://www.socialcrawl.dev/v1/search/forums?query=airpods+pro+3+battery&comments=on" \
-H "x-api-key: sc_your_api_key_here"Same key as the rest of the catalog
Every endpoint returns structured JSON in a unified envelope. Computed fields such as engagement rate and content category are included only where the endpoint supports them and the required source inputs are present.
Same Universal Search data, two very different paths. Auth, rate limits, schema, and cost — side by side.
| Feature | SocialCrawl | Per-platform integrations (DIY) |
|---|---|---|
| Platforms per call | 14 platforms (up to 17 sources in hashtag mode) with one GET | One integration per platform: 14 separate clients to write and run |
| Authentication | Single x-api-key for the whole fan-out | Separate keys, quotas, and approval queues for each platform |
| Result fusion | Weighted reciprocal rank fusion plus LLM reranking and clustering built in | Merging, dedup, and ranking logic is yours to design |
| Streaming | JSON or typed SSE chunks as each platform settles | A streaming layer has to be built on top of every client yourself |
| Pricing | Flat 20 credits per call, auto-refunded when zero results land | 14 separate bills and rate-limit ceilings to track |
| Maintenance | Every connector stays alive behind one stable contract | Each platform change breaks one of your 14 integrations |
Platforms per call
Authentication
Result fusion
Streaming
Pricing
Maintenance
Find answers to frequently asked questions about SocialCrawl's API, pricing, and capabilities.
Contact usAsk AI about SocialCrawl