At our extraction and summarisation technology is an integral part of our product. Now with our API you can use our technology in your own applications.


All of our end points are secured using an API key and should be sent via the x-api-key header to{resource}

Our API keys are 40 character alphanumeric strings and are case sensitive.

Please request your free API key above by emailing us to get started.

Requests are rate limited to 5 requests per second with bursts of up to 20 requests. If your needs exceed these limits please get in contact with us.


Skims are the complete end to end product of our extraction, identification and summarisation technology. Here is an example of what one looks like:

	"uri": ""
	"content_type": "news"
	"media_type": "image"
	"media_uri": ""
	"title": "Newcastle cancel 'inappropriate' end-of-season awards"
	"body": "Relegation-threatened Newcastle have cancelled an end-of-season awards night because of the position they are in" // ... up to around 100 words
	"markup": [
			"children": [
				{ "children": [ "Newcastle" ], "key_term": true },
				" have cancelled an end-of-season awards night because of the position they are in"
		// ... this can be quite long, again to a total of around 100 words
	"key_terms": "["
		"football", "Newcastle FC", "Relegation", "Premier League"

Skims can be created via the GET end point.

URL Parameters:

  • uri - the target uri to turn into a skim, at present this must be a publicly accessible url.
curl -H "x-api-key: ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789abcd"

This will return a JSON formatted response containing an object with an attribute skims which will be an array containing a single skim object as outlined above.

	skims: [
		{ "uri": "", ... } // a skim object

other resources

We plan to open up other sections of our API in the near future, such as separate extraction and summarisation functionality. If you have a particular use case or would just like to know more please email us for further information - we'd love to hear from you.