updated from feedback

Signed-off-by: Filinto Duran <[email protected]>

Author: filintod

conversation/python/http/README.md

@@ -19,7 +19,7 @@ This example uses the default LLM Component provided by Dapr which simply echoes
 1.  Install dependencies:
 
     <details open="true">
-    <summary>Option 2: Using uv (faster modern alternative to pip)</summary>
+    <summary>Option 1: Using uv (faster modern alternative to pip)</summary>
 
     ```
     cd conversation
@@ -35,7 +35,7 @@ This example uses the default LLM Component provided by Dapr which simply echoes
     </details>
      
     <details>
-    <summary>Option 1: Using classic pip</summary>
+    <summary>Option 2: Using classic pip</summary>
 
     <!-- STEP
     name: Install Python dependencies
@@ -59,7 +59,6 @@ This example uses the default LLM Component provided by Dapr which simply echoes
     cd ..
     ```
 
-<!-- END_STEP -->
 
 2. Open a new terminal window and run the multi app run template:
 
@@ -71,7 +70,7 @@ This example uses the default LLM Component provided by Dapr which simply echoes
       - '== APP - conversation == Tool calling input sent: What is the weather like in San Francisco in celsius?'
       - '== APP - conversation == Output message: What is the weather like in San Francisco in celsius?'
       - '== APP - conversation == Tool calls detected:'
-      - '== APP - conversation == Tool call: {'id': '0', 'function': {'name': 'get_weather', 'arguments': 'location,unit'}}'
+      - "== APP - conversation == Tool call: {'id': '0', 'function': {'name': 'get_weather', 'arguments': 'location,unit'}}"
       - '== APP - conversation == Function name: get_weather'
       - '== APP - conversation == Function arguments: location,unit'   
     expected_stderr_lines:
@@ -101,7 +100,8 @@ This example uses the default LLM Component provided by Dapr which simply echoes
     
     - The app then sends an input `What is the weather like in San Francisco in celsius?` to the `echo` Component mock LLM.
     - The mock LLM echoes `What is the weather like in San Francisco in celsius?` and calls the `get_weather` tool.
-    - Since we are using the `echo` Component mock LLM, the tool call is not executed and the LLM returns `No tool calls in response`.
+    - Since we are using the `echo` Component mock LLM, the tool call is not executed and the LLM returns but we return a simulated `tool call` response.
+      The response is mocked to showcase the tool call response and what was sent as the tool definition, like the function name and arguments.
     
     ```text
     == APP == Tool calling input sent: What is the weather like in San Francisco in celsius?
@@ -177,6 +177,9 @@ This example uses the default LLM Component provided by Dapr which simply echoes
     - The mock LLM echoes `What is dapr?`.
     - The app then sends an input `What is the weather like in San Francisco in celsius?` to the `echo` Component mock LLM.
     - The mock LLM echoes `What is the weather like in San Francisco in celsius?`
+    - The mock also "calls" the `get_weather` tool.
+    - Since we are using the `echo` Component mock LLM, the tool call is not executed and the LLM returns but we return a simulated `tool call` response.
+      The response is mocked to showcase the tool call response and what was sent as the tool definition, like the function name and arguments.
     
     ```text
     == APP - conversation == Conversation input sent: What is dapr?

conversation/python/sdk/README.md

@@ -17,7 +17,7 @@ This quickstart includes two example applications:
 
 You can run the sample applications using either the Dapr multi-app template or the Dapr CLI directly.
 
-### Option 1: Using the Multi-App Template
+### Use the multi-app-run template file
 
 This approach uses [Dapr's multi-app run template files](https://docs.dapr.io/developing-applications/local-development/multi-app-dapr-run/multi-app-overview/) to simplify deployment with `dapr run -f .`.
 
@@ -113,8 +113,8 @@ For more LLM options, see the [supported Conversation components](https://docs.d
     expected_stdout_lines:
       - "== APP - conversation-tool-calling == Input sent: calculate square root of 15"
       - "== APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='calculate square root of 15', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='calculate', arguments='expression'))]))"
-      - "== APP - conversation-tool-calling == Input sent: get weather in London in celsius"
-      - "== APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='get weather in London in celsius', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='get_weather', arguments='location,unit'))]))"
+      - "== APP - conversation-tool-calling == Input sent: get weather in San Francisco in celsius"
+      - "== APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='get weather in San Francisco in celsius', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='get_weather', arguments='location,unit'))]))"
     expected_stderr_lines:
     output_match_mode: substring
     match_order: none
@@ -133,8 +133,8 @@ For more LLM options, see the [supported Conversation components](https://docs.d
     ```text
     == APP - conversation-tool-calling == Input sent: calculate square root of 15
     == APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='calculate square root of 15', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='calculate', arguments='expression'))]))
-    == APP - conversation-tool-calling == Input sent: get weather in London in celsius
-    == APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='get weather in London in celsius', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='get_weather', arguments='location,unit'))]))   
+    == APP - conversation-tool-calling == Input sent: get weather in San Francisco in celsius
+    == APP - conversation-tool-calling == Output response: ConversationResultAlpha2Choices(finish_reason='tool_calls', index=0, message=ConversationResultAlpha2Message(content='get weather in San Francisco in celsius', tool_calls=[ConversationToolCalls(id='0', function=ConversationToolCallsOfFunction(name='get_weather', arguments='location,unit'))]))   
     ```
 
     <!-- END_STEP -->
@@ -152,7 +152,7 @@ For more LLM options, see the [supported Conversation components](https://docs.d
 
     <!-- END_STEP -->
 
-### Option 2: Using the Dapr CLI Directly
+### Run the apps individually
 
 As an alternative to the multi-app template, you can run the application directly with the Dapr CLI.
 

conversation/python/sdk/conversation/tool_calling.py

@@ -16,6 +16,11 @@
 with DaprClient() as d:
     provider_component = "echo"
 
+    # ------------------------------------------------------------
+    # Creating Tool Function definition using lower level API and hand-crafted JSON schema
+    # ------------------------------------------------------------
+
+    # This is how to register a tool using hand-crafted JSON schema
     function = conversation.ConversationToolsFunction(
         name="calculate",
         description="Perform calculations",
@@ -63,7 +68,7 @@ def get_weather(location: str, unit: str) -> str:
         """get weather from a location in the given unit"""
         return f"The weather in {location} is 25 degrees {unit}."
 
-    textInput = "get weather in London in celsius"
+    textInput = "get weather in San Francisco in celsius"
     # use create helper function (i.e.: create_user_message, create_system_message, etc...) to create inputs easily
     inputs = [
         conversation.ConversationInputAlpha2(messages=[conversation.create_user_message(textInput)], scrub_pii=True),
@@ -84,5 +89,5 @@ def get_weather(location: str, unit: str) -> str:
 
         # registered tools are also automatically set to be invoked easily when called by the LLM
         # using the method conversation.execute_registered_tool:
-        # >>> print(conversation.execute_registered_tool(name="get_weather", params={"location":"London", "unit":"celsius"}))
+        # >>> print(conversation.execute_registered_tool(name="get_weather", params={"location":"San Francisco", "unit":"celsius"}))