API for URL top tags, bookmark count

Social bookmarking site has exposed a new API providing the top tags and total number of bookmarks for any URL in its system. Yahoo’s Developer Network provided a short preview earlier tonight of a soon to be released web badge but currently anyone can request data from the open API. It’s a useful feature to provide additional context for a URL, suggest tags, or measure one aspect of a site’s popularity.


Simply submit a request to the above API endpoint with a hex MD5 hash of the URL of interest as your hash parameter value. returns results in JSON key-value pairs. Data includes the total number of users who have tagged the given URL and the top 11 tags (and tag count) used to describe its content.

You can check out a few examples such as the response for, the response for, or the response for If you need help constructing a MD5 hash you can use Paul Johnson’s implementation ( uses the same script). You may specify a callback function using the callback parameter.

The API is officially unreleased, may be shut down if not used in full Yahoo-constructed blog sidebar badge form, and may be subject to further terms of service. Hopefully the new set of servers can keep up with demand.

Update: officially announced Tagometer badges as well as a JSON feed of URL data about 16 hours after this post was published.

  • Posted
  • Updated at
  • Comments [9]


Commentary on " API for URL top tags, bookmark count":

  1. Kevin Burton on wrote:

    The real sad thing is that Delicious has historically blocked anyone from using their APIs unless you do some magic rain dance.

    … which prevented a lot of innovation.

    They blocked Technorati sporadically and they just ended giving up.

    We (Tailrank) tried to use their API early on for an experiment but they quickly blocked us. We tried to inquire the terms of the API but never received an answer.

    We might have been indexing too fast but we were never able to get a straight answer as to *exactly* how fast was too fast.

    The only way to figure out if you can really use the Delicious API is to ask the Yahoo groundhog if he saw his shadow that morning. :)

    Hopefully they’ll publish terms of service or some other details to people can actually use this thing.


  2. Joost de Valk on wrote:

    Cool stuff!

    Made a PHP implementation based on this info, quite simple actually, here it is.

  3. Justin Thorp on wrote:

    Yeah hopefully they get this API out the door soon. I’d love to play with it.

    The blogosphere/Web is so huge and there needs to be more innovation in allowing people to create communities and conversations around their content or ideas. Otherwise were just a million talking heads all talking individually.

    This API seems like it moves towards doing that.

  4. Joshua Schachter on wrote:

    This is just a simple JSON endpoint for a badge scriptlet we are going to release. Calling it an API is a overstatement and/or miscommunication. We may change formats, endpoints, etc, all the usual disclaimers apply.

  5. Adi Oltean on wrote:

    Hmm… why is this not XML?

  6. Dare Obasanjo on wrote:

    Are you going to post a correction now that has stated explicitly that the JSON endpoint is NOT an officially sanctioned API?

    • Niall Kennedy on wrote:

      I never claimed the JSON feed was an official API, just a data endpoint providing useful data in ways previously reserved for scrapers. From the JSON URL feed page:

      This feature is intended to facilitate the population of multiple widgets and display elements on a web page. The multiple record fetch is not intended as a general data query method for the purposes of harvesting data from The multiple fetch limit may be subject to change, and will be subject to usage throttling and abuse monitoring.

      FeedBurner is utilizing the JSON URL feed to generate images for inclusion in feed entries (FeedFlare). They are the first external use and were included in the announcement.

      Is the JSON URL an application programming interface? Definitely. Turn the data into your own custom web badge, place the data in an image, or you could even turn it into a screensaver. The methods of accessing data are changing, but whether it’s JSON, RSS, Atom, SOAP, or XML-RPC, etc., it’s still an interface and an envelope to me.

  7. Jonathan on wrote:

    is there a way to get user’s bookmarked urls given a tag?

  8. Ather on wrote:

    I have noticed when you get tags info after providing hash as a parameter, it ask to open target file as ‘ in what format?’. and second thing is it possible to get this in xml format. otherwise have to write php code to get these info.