Follow an incident, a policy or a public institution across social, news and the open web from one query. Every source is public, results come back on one schema, and the same call re-runs on your own cadence for as long as the story lasts.
Public sources, one query
An OSINT monitoring API collects publicly available posts, articles and pages about an event, region or public institution so an analyst can follow a story as it develops. SocialCrawl fans one query across 46 platforms covering social, news and the open web, returns everything on one shared schema, and stays inside public-source scope with no private-record access.
Every one of these runs in production today. The endpoint chain under each card is the actual call sequence.
One query with a lookback window returns ranked results from social, forums, news and the open web together. Run it on your own schedule and it becomes the collection layer for a watch desk, without a separate integration per network.
search/everywhere (standing query, lookback)Track one named subject across networks between two dates and group by day. Useful when the question is when something started rather than how loud it got, which is the question that usually matters in an after-action review.
prism/brand-mentions (date-ranged entity)Follow the posts of a specific public account on a tight cadence. This is single-account monitoring of an official or public-figure profile, not a person search, and the accounts worth watching are the ones already speaking in public.
truthsocial/user/posts (public figure)Find coverage with localised news search, then pull the page for the full text. Some publisher pages block automated readers, so keep the headline record as the durable row and treat the body text as best effort.
google_news/search -> web/scrapeSearch public events by place and keyword to reconstruct what was scheduled around a location and a date. One of the cleanest surfaces we run, and an unusual one to have through an API at all.
facebook/events/search (local incident)Run one composite pass over a subject to see whether something is escalating across networks rather than spiking on one. The cross-network shape is what separates a real incident from a single loud thread.
prism/crisis-radarFour calls cover collection, timeline and the local detail. Start with the broad sweep and drill into whichever lane returned something.
Reads are priced in credits. Empty result sets are not billed on list endpoints, and failed upstream calls are refunded automatically. Every source is public.
GET /v1/search/everywhereOne query and a lookback window returns ranked results across social, forums, news and the web on one schema. This is the collection step.
flat rate per callGET /v1/prism/brand-mentionsPass a subject and a date range to turn scattered mentions into a dated series you can plot and defend.
per callGET /v1/google_news/searchRun localised news search with language and region controls so regional coverage sits beside the social lanes rather than in a separate tool.
per callGET /v1/facebook/events/searchSearch public events by place and keyword when the question is about a specific location on a specific day.
per callEach of these is a lane in the same standing query. Follow a link to see that platform's full endpoint surface.
Most teams either buy an enterprise monitoring platform or maintain a set of scrapers. Here is what changes.
| Capability | SocialCrawl | Typical setup |
|---|---|---|
| Coverage | One key spans 46 platforms across social, news and the open web. | A social suite plus a separate news vendor plus scrapers for everything else. |
| Data scope | Public sources only, stated plainly and consistent across every platform. | Mixed sourcing that a compliance review has to unpick vendor by vendor. |
| Output | Structured records on one schema, ready to store, diff and cite with their source URLs. | A dashboard, with export as a paid add-on. |
| Timeline work | Date ranges and daily grouping are query parameters, so a dated series is one call. | Charts you can look at but not reproduce. |
| Pricing model | Never-expire credit packs, priced per call, so a short-lived watch stays a short-lived cost. | Annual platform licences sized for a permanent team. |
| Getting started | Free credits on signup and a runnable call in the first minute. | Procurement, then onboarding. |
An OSINT API returns open-source intelligence as structured data: publicly available posts, articles and web pages about a subject, event or organisation. It replaces manual collection across a dozen sites with one query that fans out and comes back on a single schema you can store and cite.
SocialCrawl covers 46 platforms and 375 endpoints, including social networks, forums, localised news, public event listings and the open web. One query can run across all of them together, or you can scope it to the lanes that matter for a given watch.
Yes. Everything returned is publicly visible, exactly as any visitor to that page or profile would see it. There is no private-record access, no data behind a login, no purchased identity data and no de-anonymisation anywhere in the product.
The API is built for public figures, official accounts, organisations, places and topics. It exposes no private records, no contact details and no identity graph on private individuals, so a person-targeting workflow is not something it can serve. Public-figure and institutional monitoring is the supported shape.
SocialCrawl returns public data only, which is the basis on which customers with a European compliance load use it. You remain the controller for whatever you store and for the lawful basis of your own processing, so document the purpose, keep retention tight and scope collection to it. We can share our data handling position on request.
Yes, by re-running your query from your own scheduler. Most production monitoring on the API is a cron job, an n8n workflow or an edge function polling a fixed query set and storing what is new. Repeat lookups of the same subject are the most common production pattern here.
Reads are priced in credits from never-expire packs, so cost tracks the number of calls you run rather than a seat count or a channel count. A short watch costs what a short watch should. Signup includes 100 free credits and no card is required.
Ask AI about SocialCrawl
100 free credits on signup. Public sources only, no card, no procurement cycle.
curl -G "https://socialcrawl.dev/v1/search/everywhere" \
-H "x-api-key: $SOCIALCRAWL_API_KEY" \
--data-urlencode "query=port strike supply disruption" \
--data-urlencode "lookback_days=3"curl -G "https://socialcrawl.dev/v1/prism/brand-mentions" \
-H "x-api-key: $SOCIALCRAWL_API_KEY" \
--data-urlencode "keyword=municipal budget vote" \
--data-urlencode "date_from=2026-07-01" \
--data-urlencode "date_to=2026-08-01"Both calls run against live data. Swap the key and the query and they work as written.