Pass two directories into the test runner

docs/interface.md

@@ -5,10 +5,11 @@ All interactions with the Exercism website are handled automatically. Test runne
 ## Execution
 
 - A test runner should provide an executable script. You can find more information in the [docker.md](https://github.com/exercism/automated-mentoring-support/blob/master/docs/docker.md) file.
-- The script will receive two parameters:
+- The script will receive three parameters:
   - The slug of the exercise (e.g. `two-fer`).
-  - A path to a directory containing the submitted solution file(s) and the necessary test file(s) (with a trailing slash).
-- The script must write a file to that directory named `results.json`
+  - A path to an input directory containing the submitted solution file(s) and the necessary test file(s) (with a trailing slash).
+  - A path to an output directory (with a trailing slash).
+- The script must write a file to the output directory named `results.json`
 
 ## Output format
 
@@ -40,7 +41,6 @@ The following per-test statuses are valid:
 - `fail`: The test failed
 - `error`: The test errored
 
-
 The per-test `message` key can be used to display human-readable error messages. Presume that whatever is written here will be displayed to the student. If there is no error message, either set the value to `null` or omit the key entirely.
 
 **We will provide a Ruby script that converts JUnit to this JSON output, which you can add as a final step of your Docker images, if you prefer to use an existing format.**
@@ -49,6 +49,6 @@ The per-test `message` key can be used to display human-readable error messages.
 
 The contents of `stdout` and `stderr` from each run will be stored in files that can be viewed later.
 
-You may write an `results.out` file that contains debugging information you want to later view.
+You may write an `results.out` file to the output directory, which that contains debugging information you want to later view.
 
 At a later date, we will provide an interface for you to download these files along with the submitted files and `results.json`.