Scrape a url. server would immediately reply by:

{ id: <url-id> }

If this is the first time this url was called, it would start scraping it asynchronously. The status code would be 201.
Otherwise, it would just return status 200.

If you wish to force-scrape it anyway, pass 'force=true':
POST /stories?force=true&url={url}

Get details for a previously scraped url.

• If url was never scraped, it would return status 404.
• If url is currently being scraped, it would return:

{
  "id": <url-id>
  "scrape_status": "pending",
  "updated_time": <ISO8601 time>
}

• If scraping had failed, it would return:

{
  "id": <url-id>
  "scrape_status": "error",
  "updated_time": <ISO8601 time>
}

Otherwise, it would return the scraping result:

{
  "scrape_status": "done",
  "id": "5ad42c6c7b4990e6ceeb326006ec9950",
  "title": "Open Graph protocol",
  "type": "website",
  "url": "http://ogp.me/",
  "image": {
      "url": "http://ogp.me/logo.png",
      "type": "image/png",
      "width": "300",
      "height": "300",
      "alt": "The Open Graph logo"
  },
  "description": "The Open Graph protocol enables any web page to become a rich object in a social graph.",
  "updated_time": "2018-06-19T09:34:20.182Z"
}

Note: image, or any other properties would be an array if there is more than one tags by that name.

Create scrape:
curl -X POST /stories?url=http%3A%2F%2Fogp.me%2F

Then get the created scrape:
curl -X GET /stories/5ad42c6c7b4990e6ceeb326006ec9950