31 actors live on Apify

Public records in, clean JSON out.

Ready-to-run APIs that turn public data into structured results: FDA recalls, federal grants, court opinions, SEC filings, provider and broker lookups, security advisories and more. Run one on demand, pay $0.007 an event, and skip the scraper you were about to write.

  • No subscription
  • No proxies, no logins
  • Source on GitHub

Live demo$0.007/event

Website Email & Phone Extractor for Lead Gen

  • Runs in your browser
  • Free account
  • No card
Try it now

Actors live

31

Per event

$0.007

Subscriptions

Zero

A real catalog entry and live portfolio numbers, not a mock-up.

Public data, played straight

Official public APIs and open government data. No proxies, nothing that works around bot protection, and no actor logs in to anything or asks for your passwords. When a page has to be read directly, robots.txt is respected.

You see the price first

Every actor costs $0.007 per event, $7 per 1,000. No subscription, no seats, no minimum. Many charge once per search rather than per row, so a search returning 50 results still costs $0.007.

Source included

Every actor ships with its code on GitHub. Read how it works, fork it, or check exactly which endpoints it touches before you spend anything.

Built in public

A running log of what shipped, what broke, and what the data source did that its own docs never mentioned.

How it works

Four steps, then you have the data.

There is nothing to install and no server to keep alive. Pick a feed, see what it returns, and run it when you need it.

  1. Pick a feed

    Browse the catalog by what you actually need: recalls, grants, filings, court opinions, security advisories, weather. Each one lists its source and its price up front.

  2. Try it in the browser

    Most feeds have a live demo page. Run a real query against the real source, see the real shape of the data, and spend nothing doing it.

  3. Run it on Apify

    Open the actor on Apify and run it with your own input, on demand or on a schedule. Nothing to install and no server of your own to keep alive.

  4. Take the JSON

    Results land in a dataset you can read over the API or export as JSON, CSV or Excel. Wire it into whatever you were going to build anyway.

One thing to expect: these are small, single-purpose tools, not a managed data platform. They fetch from the source at run time, so an actor is only ever as fresh and as available as the public source behind it.

The catalog

Pick a feed, or see what is next.

8 featured actors out of 31 live, each one calling an official public source and returning structured results. 6 of them have a live demo you can run right here.

GET /website-lead-extractor

Website Email & Phone Extractor for Lead Gen

Crawl a company website and extract publicly listed emails, phone numbers, and social profile links (LinkedIn, X/Twitter, Facebook, Instagram, GitHub) as clean JSON.

Try it live →
GET /website-tech-stack-detector

Website Tech Stack Detector: CMS, Hosting & Analytics

Enter website URLs and get each site's tech stack in one row: CMS, ecommerce platform, JS framework, analytics, CDN/hosting, payment and chat tools, plus server headers.

Try it live →
GET /certificate-transparency-monitor

SSL Certificate Monitor: New Certs & Subdomains

Search Certificate Transparency logs for newly issued SSL/TLS certificates on a domain and its subdomains. Spot unauthorized certs or newly launched subdomains — official CT log data, no proxy, no scanning.

Try it live →
GET /field-operations-risk-briefing

Field Ops Risk Briefing: Weather & GPS Conditions

A go/no-go risk briefing for U.S. field operations: weather forecast, active weather alerts, and GPS and HF radio conditions (Kp index, radio blackout level, aurora line) for each work site in one row. For survey, drone, agriculture, utility and field service teams.

Try it live →
GET /federal-contract-award-tracker

Federal Contract Awards: GovCon Leads & Intel

Search recent U.S. federal contract awards by keyword, NAICS code, agency, or state. Get back who just won, how much, and from which agency — for subcontracting leads and GovCon competitor tracking.

Try it live →
GET /us-weather-tracker

US Weather Forecast & Alerts API (NWS)

A go/no-go weather check for U.S. locations: official National Weather Service forecast and active severe weather alerts for each site. For event planners, delivery and logistics dispatch, and construction and site managers deciding whether to go ahead.

Try it live →
GET /sec-filing-full-text-search

SEC Full-Text Search: Any Filing, Any Keyword

Search a keyword across every SEC filing type at once (10-K, 10-Q, 8-K, S-1, and more), not just one form. Get back matching filings with company, form type, item codes, and a direct link. No key, no proxy.

GET /npi-registry-lookup

NPI Lookup: Verify Healthcare Providers

Look up US healthcare providers and organizations by NPI number, name, or location: credential, taxonomy/specialty, license, status, and address, via the official CMS NPPES NPI Registry API.

What it costs

One price, every actor.

Billing is per event on Apify: you pay for the runs you make and nothing else. No plan to pick, no seats to count, no minimum spend.

Pay per event

All actors

$0.007

per charged event — $7 per 1,000

Subscription
None
Seats
None
Minimum spend
None

Live demos

On this site

Free

a few runs a day, shared by everyone

Account needed
Free one
Card needed
No
Demos available
6

How the billing works

  • A charged event is defined per actor and named in its README. Many actors charge once per search, so one query that returns fifty rows is still one event.
  • Billing and payment are handled by Apify against your own account. Nothing is charged on this site and no card details are collected here.
  • A run that fails before it charges costs nothing. You are not billed for a source being down.
  • Every actor page on Apify shows its live pricing. If a price ever changes, that page is the record.
Straight answers

What this is, and what it is not.

These are small, sharp tools for getting public data out of public sources. They are good at that and deliberately bad at everything else.

You can expect

  • Structured JSON from sources that publish it openly
  • A price you can see before you run anything
  • Code on GitHub for every actor in the catalog
  • Live demo pages for most of the catalog

Please do not expect

  • Data from behind a login, a paywall or a bot wall
  • Scraped personal data or anything a source forbids
  • A monthly subscription or a per-seat plan
  • An SLA — these are small tools, run as such
FAQ

Questions people ask first

What does it actually cost?

Every actor in the portfolio charges $0.007 per event, which works out to $7 per 1,000. There is no subscription and no seat cost. Many actors charge once per search rather than once per result row, so a query that returns fifty records still costs a single $0.007 event.

Do I need an account to try one?

A free one, for the live demos. The demo pages run a real query against the real data source, which costs us a little each time, so an account keeps that honest and gives the daily cap something real to count. No card is asked for. To run an actor with your own input or on a schedule you will need an Apify account instead, because that is the platform the actors run on.

Where does the data come from?

Official public APIs and open government data: the FDA, FEMA, USAspending, the SEC, CourtListener, the NWS, USGS, GitHub Security Advisories, and similar sources. Each actor names its source in its README and on its Apify page.

Do these scrape sites or use proxies?

No. Nothing in the portfolio uses proxies, works around bot protection, or logs in to anything. Where an actor reads a page directly rather than calling an API, it checks robots.txt first. Any source that would need an account or a key the user has to create is left out of the catalog.

Can I see the code?

Yes, all of it. Every actor has a public GitHub repository under the same name as its Apify slug, linked from its card in the catalog.

How fresh is the data?

Each run fetches from the source at the moment you run it, so the data is as current as the source makes it. How often a source itself updates varies: some publish continuously, others on a daily or weekly schedule.

Can you build one for a source I need?

Often, yes, as long as the source is public and does not require an account, a proxy or anything that works around bot protection. The build log tracks what is shipping next and the GitHub profile is the quickest way to get in touch.

Have a specific feed in mind?

Every actor here started as someone’s specific need. Read the source, run a demo, or follow the build log for what is shipping next.

Public sources only · $0.007 per event · No proxies, no logins