Download the PHP package aporat/scrapecreators-php without Composer

Scrape Creators PHP

A PHP client for the Scrape Creators API: a typed method for every documented endpoint across TikTok, Instagram, YouTube, X, Reddit, LinkedIn, Facebook, GitHub, Threads, Bluesky, Spotify and twenty more platforms, control of the provider-side cache that turns repeat lookups into zero-credit calls, and exceptions that tell a missing profile from a restricted one from an outage, including the HTTP 200 placeholders the API uses for some of them. Works in any PHP project; a service provider is included for Laravel.

Not affiliated with Scrape Creators.

✨ Features

• Every endpoint, typed. 251 methods on 27 platform resources, generated from the API's OpenAPI document: required parameters are positional, optional ones are named, each carries the API's own description and allowed values. $client->instagram()->profile('nasa', trim: true). Anything newer than this package is reachable through get() / post() or the $extra parameter every method takes
• Provider-side cache, handled. Set a default cache_max_age once and the client sends it to every endpoint that supports it; Scrape Creators then answers from a recent scrape for 0 credits. Pass Client::LIVE on a call to force a fresh scrape
• Errors you can branch on. NotFoundException for a missing account, RestrictedException for a private, age-gated or region-blocked one (with isAgeRestricted() and isPrivate()), UnavailableException for their outage or a timeout, InsufficientCreditsException, AuthenticationException, InvalidRequestException, RateLimitException. The API reports several of these states as HTTP 200 placeholders; the client reads the envelope so you do not have to
• Billing in view. Every Response exposes creditsCharged(), creditsRemaining(), isCached() and cachedAt() alongside data() and dot-path get()
• Laravel ready. Auto-discovered service provider, config/scrapecreators.php, Client bound as a singleton

🛠 Installation

Requires PHP 8.4+ and Guzzle 7.9 or 8. Get an API key at app.scrapecreators.com.

📚 Usage

Platforms

One accessor per platform, one method per endpoint. Method names follow the path: /v1/tiktok/video/transcript is $client->tiktok()->videoTranscript($url), /v3/tiktok/profile/videos is profileVideos(), /v1/account/get-api-usage is $client->account()->apiUsage().

The three endpoints that also accept a JSON body have a ...Post() twin taking the body as an array: facebook()->adLibrarySearchAdsPost([...]), facebook()->adLibraryCompanyAdsPost([...]), reddit()->postCommentsPost([...]).

Any endpoint

Booleans are sent as true / false, nulls are dropped, enums are sent by value. cacheable says whether the endpoint takes cache_max_age, so the client's default applies.

Responses

The provider-side cache

Thirty-six endpoints take cache_max_age (1d, 3d, 7d, 14d, 30d): when Scrape Creators holds a scrape at least that fresh it returns it for 0 credits, marked cached: true with cached_at. Set a default once and override per call:

A cached profile carries whatever signed CDN image URLs the scrape saw, and networks expire those within days. Keep the default short if you download pictures.

Errors

How responses map:

🧱 Laravel

The service provider is auto-discovered and binds Aporat\ScrapeCreators\Client as a singleton. Publish the config and set the key:

In tests, bind a client built on a Guzzle MockHandler, or $this->app->instance(Client::class, ...).

🔁 Keeping up with the API

spec/openapi.json is a snapshot of docs.scrapecreators.com/openapi.json. src/Resources is generated from it:

CI fails when the committed resources differ from what the snapshot produces. Until a new endpoint lands here, call it with get() / post().

📄 License

MIT. See LICENSE.