Getting Started
ScreenshotAPI is a simple API that allows you to take programmatic screenshots of any website with ease. Screenshots can be saved as JPEG, PNG, WebP, and even PDF! We provide plenty of options.
Last updated
ScreenshotAPI is a simple API that allows you to take programmatic screenshots of any website with ease. Screenshots can be saved as JPEG, PNG, WebP, and even PDF! We provide plenty of options.
Last updated
The following options for the API call are:
TOKEN: The TOKEN
or your API key should be replaced by your real ScreenshotAPI API, you can find your API key after you Sign Up or Log In to the service via the Dashboard page.
Note: If for any reason you need to get a new API key, simply click "Roll API Key" in the "Settings" card on the Dashboard. This will issue you a fresh API key and revoke access from the previous key.
URL: Website URL for the site you would like to render for the screenshot. We support a wide variety of websites types, such as single page apps, very long-content, and sites that require lazy loading or delays ensuring accurate and complete screenshots of these unique sites.
OPTIONS: Options indicates the different URL parameters keys and values that can be used to render your "perfect" screenshot. All the options for screenshots will be detailed below.
Note: Default options for the screenshot will be shown in the Options table below.
Example:
This will render the following png
image output when you paste in the above string into your browsers URL bar (replacing TOKEN with your API key):
When you take a screenshot, it automatically get cached on our backend, so when you make a get request to the same URL this will return the same screenshot that got cached, but if you want to take a newer screenshot of the URL provided, you may want to use our fresh screenshot functionality.
Example:
In the following video, you will see an example of how the "Fresh Screenshot" functionality is used. First, we demonstrate how a regular request to a specific URL is made for the first time, which typically takes a few seconds since it's the initial request. Then, the same request is made a second time, which is much faster because it retrieves the cached screenshot. However, if you want to capture a completely new screenshot without using the cached one, you should click the checkbox labeled 'Fresh screenshot' as shown in the example during the third request in the video.
You can select elements by class name, ID, or other CSS selectors and then add the display: none
attribute to hide them on the inject CSS feature.
Example:
In the following video, you'll find a more illustrative example of how to use the 'Inject CSS' functionality to remove an element from a page. In the first request, you'll see a standard screenshot to observe the page's initial state. Then, in the second request, we demonstrate how to achieve this by utilizing the 'Inject CSS' feature to remove the desired element.
You can access the 'Scrolling Screenshot' functionality, which allows you to capture a video (in webm, gif, or mp4 format) of a web page with various customizable options. These options include the ability to scroll backward, adjust scrolling speed, set the duration, and start recording instantly
Example:
In the following video, you'll find a more illustrative example of how to utilize the 'Scrolling Screenshot' functionality. When you click the 'Take Screenshot' button, it will take a few seconds to capture the video and provide you with JSON containing the video information.
Here is the recorded video demonstrating the example.
In addition to the 'Scrolling Screenshot' functionality, our application offers the feature of 'Multiple Scrolling Screenshots.' With this feature, you can capture multiple scrolling screenshots of web pages, each with different sizes in terms of width and height. This means you have the flexibility to customize the dimensions of the screenshots according to your specific requirements.
Example:
In the parameters of the endpoint, we will pass an array of objects with different sizes. To achieve this, we need some code to pass this array in the 'sizes' parameter.
In the following video, you'll find a more illustrative example of how to utilize the 'Multiple Scrolling Screenshot' functionality. When you click the 'Take Screenshot' button, it will take a few seconds to capture the video and provide you with JSON containing the video information.
POST mode accepts the exact same parameters in the request body, but you need to use a JSON.
GET
https://shot.screenshotapi.net/screenshot?token=TOKEN&url=https://apple.com
GET
https://shot.screenshotapi.net/screenshot?token=TOKEN&url=https://apple.com&scrolling_screenshot=true
In rare circumstances the API may encounter an "Error 524" from Cloudflare. This is due to a timeout between the ScreenshotAPI API and our servers due to auto-scaling. If you see this error please just try your request again.
You may also receive an error if you select the no_cookies_banner
as TRUE and try to export as a PDF. There are some limitations that do not allow these two settings to happen at the same time.
With each API call the following quota headers are sent.
The total number of screenshots you can take per month
Use our Query Builder to explore our API. (If you are logged in your API key is automatically added). To easily view all the screenshot options in an interactive builder to ensure you have your ideal screenshot dialed in.
You can check out the Query Builder here: https://app.screenshotapi.net.
Inject Javascript field is a javascript code that the browser can execute before taking the screenshot.
For example: Try taking a screenshot of https://xo.capital/ and paste this code in the "Inject JS code" input of the query builder:
document.getElementsByClassName('display-heading')[0].innerHTML = 'This is a test to help the customer'
We added a new option recently specifically for the PDF output. You can select the range that you want to select. Sometimes if the screenshot has unnecessary details that bleed into another page and you want to remove that from the screenshot, you can customize the page range that gets outputted directly from the Query Builder.
This option on the query builder allows you to store the HTML and the text of the site. You currently need to set the output as json to grab the URLs. If you are using the img output, for example, we will extract the information but there is no way to grab the URLs.
Below is sample code for the following languages: Node.JS, PHP, Go, Java, Python, and Ruby.
If you need assistance integrating into our API with a given language please contact us.
We have a free proxy you can use if you need one: http://xo:llksoigsgrewvc@137.184.64.168:3128
At ScreenshotAPI we are there for you. We are developers ourselves and strive to ensure you have a great screenshot experience.
If the ScreenshotAPI.net API is unavailable to all for more than three consecutive hours in any one calendar day, beginning at 12:00:01 AM PST and ending at 11:59:59 PM PST time of the same day, an outage will be deemed to have occurred. If there are more than three outages within any one calendar week, beginning at 12:00:01 AM PST on a Monday and ending at 11:59:59 PM PST on a Sunday, then You may terminate this Agreement, without liability, by providing written notice to ScreenshotAPI.net of Your intention to do so within one calendar week of the third outage. Upon such termination being accepted by ScreenshotAPI.net, You shall be entitled to a refund equal to the amount You have paid for the latest term.
If you need help at any point please contact us at info@screenshotapi.net or go to help.screenshotapi.net
Name | Type | Description |
---|---|---|
Name | Type | Description |
---|---|---|
file_type
string
for the file (If output
is not set to json
), the options include PNG
, JPG
, WebP
, and PDF
. The default is PNG
omit_background
boolean
Omit background (If using PNG as the file_type only) removes the background from websites with a basic white background and makes it transparent. Default is false
.
destroy_screenshot
boolean
If true
, the screenshot will not be stored on our server after render. The resulting screenshot (or PDF) must be deleted (destroyed) when it's rendered as it will not be available again. Default is false
fail_on_error
boolean
If fail on error is set to true, then the API will return an error if the render encounters a 4xx
or 5xx
status code. Default is false
longitude
number
The longitude to use with the browser's Geolocation API. Default is null
latitude
number
The latitude to use with the browser's Geolocation API. Default is null
proxy
string
Use an address for a proxy server for the request. Example: address:port
or username:password@address:port
. Default: ''
no_cookie_banners
string
Block or hide cookie banners on websites as the request loads before render. Default is false
block_ads
boolean
Block ad requests from common and popular ad-networks from loading the request before the render. Default is false
headers
string
Sets the header or headers when the request is being loaded before the render. This works for single or multiple headers. Example: X-HEADER: value; X-OTHER_HEADERL otherValue;
. Default is ''
cookies
string
Sets the cookie or cookies when the request is being loaded before the render. This works for single or multiple cookies. Example: cookie=value;otherCookie=otherValue;
. Default is ''
scroll_to_element
string
Target a specific element for the browser to scroll to before the render. This is useful if a given element is only loaded in the viewport. Default is ''
selector
string
Specify the target for the render based on a element with a matching selector. If the element is not found, a render of the results is still returned. Example: div > .main-navigation > .logo
. Default is ''
css
string
CSS code injected into the render. Default is ''
css_url
string
URL for CSS code injected into the render. Default is ''
ttl
integer
TTL (Time to Live) sets the number of seconds to keep a rendered screenshot (or PDF) in the cache. Default is 30 days or 2592000
seconds
user_agent
string
Sets the User-Agent string for the render for a particular request. Default is ''
accept_languages
string
Sets the Accept-Language header on the request for the specified URL rendered. Default isen-US,en;q=0.8
delay
integer
Time delay in milliseconds (ms) before the screenshot is rendered from the browser instance (this includes PDFs). Default is 0
thumbnail_width
integer
The width in pixels of the thumbnail generated from the render. If not set, the fallback behavior is the outputted screenshot. Default is null
output
string
Set the output of the results from the render. The output can be either JSON
or the raw image captured. Default is image
fresh
string
Take a fresh screenshot or render (or PDF) vs. getting the version cached version within the app's storage. Default is false
lazy_load
boolean
If lazy load is set to true
, the browser will cross down the entire page to ensure all content is loaded in the render. Default is false
full_page
boolean
Capture the full page of a website vs. the scrollable area that is visible in the viewport upon render. Default is false
retina
boolean
Retina or a high definition equivalent for a device that sets the pixel ratio to 2X. This option will cause the screenshot processing time to go up due to processing of the larger images. Default is false
height
integer
Viewport height in pixels of the browser render. Default is 867
width
integer
Viewport width in pixels of the browser render. Default is 1680
custom_html
string
Provide custom HTML that should be rendered. This will override the URL option and take a screenshot of the following HTML output. Default is ''
url
string
The URL of the website for the screenshot / render
token
string
Your API key for Screenshot API
file_type
string
for the file the options include WEBM
, MP4
and GIF. The default is WEBM
scroll_speed
string
this field stands for the scrolling speed, the options are fast
normal
and slow
. The default is normal
duration
integer
this field controls the scrolling duration, which in turn affects the video duration in seconds. You can enter a value between 0
and 100
.
scroll_back
boolean
if this field is set to true
, the scrolling will go to the end of the page, and once it's done, it will return to the start of the page.