Extracts the data you need from any web page by using AI. Use query schema to guide the output either with a plain instruction or with a schema like JSON string.
Extracts the data you need from any web page by using AI. Use query_schema to guide the output either with a plain instruction or with a schema-like JSON string.
Endpoint: GET https://api.outscraper.cloud/ai-scraper
| Name | In | Type | Required | Description | Example | Constraints |
|---|---|---|---|---|---|---|
query |
query | array<string> | Yes | Links to web pages (e.g., https://www.apple.com/iphone/).
It supports batching by sending arrays with up to 1000 queries (e.g., query=text1&query=text2&query=text3). It allows multiple queries to be sent in one request and to save on network latency time. |
- | - |
query_schema |
query | string | No | Guides the JSON extraction output for each page.
It can be either a plain prompt (e.g., Extract company_name, company_description, and people.) or a schema-like JSON string (e.g., {"type":"object","required":[],"properties":{"company_name":{"type":"string"},"company_description":{"type":"string"},"people":{"type":"array","items":{"type":"string"}}}}).
When passing JSON, send it as a string and URL-encode the value. Maximum length: 2024 characters. |
Extract company_name, company_description, and people. |
max length 2024, default Extract base information from page. |
limit |
query | integer | No | The parameter specifies the limit of pages to get from one query. | - | default 1 |
fields |
query | string | No | The parameter defines which fields you want to include with each item returned in the response. By default, it returns all fields. Use &fields=query,name to return only the specific ones. |
- | - |
async |
query | boolean | Yes | The parameter defines the way you want to submit your task to Outscraper. It can be set to false to open an HTTP connection and keep it open until you got your results, or true (default) to just submit your requests to Outscraper and retrieve them later (usually within 1-3 minutes) with the Request Results endpoint. Each response is available for 4 hours after a request has been completed.
A good practice is to send async requests and start checking the results after the estimated execution time. Check out this Python implementation as an example.
As most of the requests take some time to be executed the async=true option is preferred to avoid HTTP requests timeouts. |
- | default true |
ui |
query | boolean | No | The parameter defines whether a task will be executed as a UI task. This is commonly used when you want to create a regular platform task with API.
Using this parameter overwrites the async parameter to true. |
- | default false |
format |
query | array<string enum> | No | The parameter defines the format of the output data. Using this parameter overwrites the ui parameter to true. |
- | - |
webhook |
query | string | No | The parameter defines the URL address (callback) to which Outscraper will create a POST request with a JSON body once a task/request is finished. Using this parameter overwrites the webhook from integrations. | - | - |
The response contains the status of the request and data. Data is an array where each element represents a response for a single query from the request.
{
"id": "your-request-id",
"status": "Success",
"data": [
{
"query": "https://www.apple.com/",
"company_name": "Apple",
"company_description": "Apple designs consumer electronics, software, and online services.",
"people": [
"Tim Cook",
"John Ternus"
]
}
]
}The response contains a request ID that can be used to fetch results by using Request Results endpoint. Each response is available for 4 hours after a request has been completed.
{
"id": "your-request-id",
"status": "Pending",
"results_location": "https://api.outscraper.cloud/requests/your-request-id"
}Wrong or missing API Key (token).
{
"error": true,
"errorMessage": "401 Unauthorized: The server could not verify that you are authorized to access the URL requested. You either supplied the wrong credentials (e.g. a bad password), or your browser doesn't understand how to supply the credentials required."
}Past due invoices or a payment method not connected.
{
"error": true,
"errorMessage": "Please close past due invoices or verify your card information is correct."
}Wrong query url parameters.
{
"error": true,
"errorMessage": "422 Unprocessable Entity: The request was well-formed but was unable to be followed due to semantic errors."
}curl -G "https://api.outscraper.cloud/ai-scraper" --data-urlencode "query=https://www.apple.com/" --data-urlencode "query_schema=Extract company_name, company_description, and people." --data-urlencode "async=false" -H "X-API-KEY: YOUR-API-KEY"curl -G "https://api.outscraper.cloud/ai-scraper" --data-urlencode "query=https://www.apple.com/" --data-urlencode "query_schema={\"type\":\"object\",\"required\":[],\"properties\":{\"company_name\":{\"type\":\"string\"},\"company_description\":{\"type\":\"string\"},\"people\":{\"type\":\"array\",\"items\":{\"type\":\"string\"}}}}" --data-urlencode "async=false" -H "X-API-KEY: YOUR-API-KEY"Finds company firmographics such as revenue, size, founding year, public status, etc.
Translates human readable addresses into locations on the map (latitude, longitude).
Translate locations on the map into human readable addresses.
Captures high quality screenshots of any webpage simply by providing its URL
Extracts the data you need from any web page by using AI.