diff --git a/src/bmm/workflows/bmad-quick-flow/quick-spec/steps/step-04-review.md b/src/bmm/workflows/bmad-quick-flow/quick-spec/steps/step-04-review.md
index ae8ee8c3f..4088c9910 100644
--- a/src/bmm/workflows/bmad-quick-flow/quick-spec/steps/step-04-review.md
+++ b/src/bmm/workflows/bmad-quick-flow/quick-spec/steps/step-04-review.md
@@ -124,18 +124,20 @@ Saved to: {finalFile}
 
 Once you are fully satisfied with the spec (ideally after **Adversarial Review** and maybe a few rounds of **Advanced Elicitation**), it is recommended to run implementation in a FRESH CONTEXT for best results.
 
-To run **Adversarial Review** in a fresh context (recommended for information asymmetry):
+Load and follow **Adversarial Review** in a fresh context (recommended for information asymmetry):
 
 \`\`\`
-/bmad-review-adversarial-general {finalFile}
+/bmad-review-adversarial-general "{finalFile}"
 \`\`\`
 
-To start **development** in a fresh context:
+Load and follow **development** in a fresh context:
 
 \`\`\`
-quick-dev {finalFile}
+/quick-dev "{finalFile}"
 \`\`\`
 
+_(Both are slash commands — prefix `/` invokes a BMAD skill or workflow.)_
+
 This ensures the agent has clean context focused solely on its task.
 ```
 
@@ -185,7 +187,7 @@ b) **HALT and wait for user selection.**
 When you're ready to implement, run:
 
 ```
-quick-dev {finalFile}
+/quick-dev {finalFile}
 ```
 
 Ship it!"