diff --git a/README.md b/README.md
index 483ef8e..f7c5a01 100644
--- a/README.md
+++ b/README.md
@@ -43,7 +43,11 @@ For more details on [supported browsers and troubleshooting](https://docs.micros
 
 Your team should also be familiar with the following:
 
-- [Forking](https://docs.github.com/github/getting-started-with-github/quickstart/fork-a-repo) and [cloning](https://docs.github.com/github/creating-cloning-and-archiving-repositories/cloning-a-repository-from-github/cloning-a-repository) repositories
+- [Forking](https://docs.github.com/github/getting-started-with-github/quickstart/fork-a-repo) 
+
+- [Cloning](https://docs.github.com/github/creating-cloning-and-archiving-repositories/cloning-a-repository-from-github/cloning-a-repository) 
+
+- If you have not used git before we suggest [GitHub Desktop](https://desktop.github.com/)
 
 ## Resources
 
diff --git a/goals/back-end.md b/goals/back-end.md
index 376fa29..6ab268e 100644
--- a/goals/back-end.md
+++ b/goals/back-end.md
@@ -113,6 +113,10 @@ Your team might find these resources helpful:
 
 - To return a value, you need to use the *Respond to a PowerApp or flow* action. Respond with text, passing the first prediction's `tagName` in a text field called `breed`.
 
+- So body("Parse_JSON") takes the result JSON as a dynamic object, which contains an array of predictions. The first() part gets the first item from this array, and the .tagName extracts the tag name field from that array item `tagname` = customvision.ai tag name of the image and tag = `breed` of dog
+
+- Ensure you add this as Expression and not Dynamic Content
+
 ## Final result
 
 ![A power automate flow showing 4 blogs in a sequence - Power Apps, classify an image, Parse JSON and Respond to a power app or flow](./media/power-automate-flow.png)
diff --git a/model-images/training-images/golden-doodle-5.jfif b/model-images/training-images/golden-doodle-5.jfif
new file mode 100644
index 0000000..f0458dc
Binary files /dev/null and b/model-images/training-images/golden-doodle-5.jfif differ