You can create as many workers as you'd like per work unit and each worker is private only to you and isolated from other workers. You can also customize worker names to reflect how you plan on using them.
Running a worker requires an active server to perform HTTP requests with. You'll be asked to launch a server if you don't already have one.
Workers execute HTTP requests included in their work unit's tasks. If the work unit is a task itself, the worker will simply execute that task's HTTP request. If the work unit is a workflow, the worker will perform as many task HTTP requests as needed to fully execute the workflow.
You may perform ad hoc runs through this website or API. Ad hoc runs will execute immediately and return the raw HTTP response if using the website.
Workers may also be scheduled to run automatically without supervision, launching and terminating servers as needed to use the least amount of credits.
Some work units declare default values for certain inputs, meaning that if you don't provide a value then the default will be used instead.
Some inputs may be difficult to find - such as client key used over SSL in an iPhone app. For this reason, you can connect your iPhone to an active server's proxy and automatically "sniff" the value to use.
If you plan on running a worker multiple times, odds are you don't want to have to keep entering in the same inputs each time. The vault allows you to save values for each parameter to use on each worker run.
Sometimes you'll want to store sensitive values to the vault, such as passwords. To add an extra layer of security, you can save these encrypted in the vault. This means you will never see the value again after you enter it (unless you enter your password).
If the tasks in a worker's work unit have collections, these may be saved as collection results on each run.
You may download collection results on the website or via API in JSON format, allowing you to use aggregate structured data from your worker's history.
If you no longer want your worker, you can archive it. This will permanently delete all collection results and schedules, but retain any associated vault entries and request logs.
Every time a worker is run, we record the following metadata about each request:
We do not save parts of the request where sensitive data may be (headers, request bodies, response bodies, etc...).