Update prompts.yaml

prompts.yaml

@@ -9,6 +9,13 @@ system_prompt: |-
   These print outputs will then appear in the 'Observation:' field, which will be available as input for the next step.
   In the end you have to return a final answer using the `final_answer` tool.
 
+  If after step 8 you have not yet called `final_answer(...)`, you **must** call:
+    ```py
+    answer = rag_fallback(question=question)
+    final_answer(answer)
+    ```  
+  and then stop.
+
   Here are a few examples using notional tools:
   ---
   Task: "Generate an image of the oldest person in this document."
@@ -319,8 +326,12 @@ managed_agent:
     Here is the final answer from your managed agent '{{ name }}':
     {{ final_answer }}
 
-fallbacks:
-  "mercedes sosa": "3"
+# -------------------------------------------------------
+# This is your *hard* RAG lookup map – ensure the key
+# here is exactly `rag_fallbacks:` with no indentation
+# -------------------------------------------------------
+rag_fallbacks:
+  "mercedes sosa": "2"
   "highest number of bird species": "Unanswered"
   "reverse prompt": "Right"
   "chess position": "Unanswered"