Addressed comments

Author: gavin-aguiar

azure/functions/decorators/function_app.py

@@ -1445,7 +1445,10 @@ def decorator():
 
     def assistant_skill_trigger(self,
                                 arg_name: str,
-                                task_description: str,
+                                function_description: str,
+                                function_name : Optional[str] = None,
+                                parameter_description_json: Optional[str] = None,  # NoQA
+                                model: Optional[OpenAIModels] = OpenAIModels.DefaultChatModel,  # NoQA
                                 data_type: Optional[
                                     Union[DataType, str]] = None,
                                 **kwargs: Any) -> Callable[..., Any]:
@@ -1459,16 +1462,22 @@ def assistant_skill_trigger(self,
         You can define functions that can be triggered by assistants by using
 
         the `assistantSkillTrigger` trigger binding. These functions are
-        invoked by the extension when a assistant signals that it would like
+        invoked by the extension when an assistant signals that it would like
         to invoke a function in response to a user prompt.
 
         The name of the function, the description provided by the trigger,
         and the parameter name are all hints that the underlying language model
         use to determine when and how to invoke an assistant function.
 
-        :param arg_name: The name of the variable that represents
+        :param arg_name: The name of trigger parameter in the function code.
         :param function_description: The description of the assistant function,
          which is provided to the model.
+        :param function_name: The assistant function, which is provided to the
+        LLM.
+        :param parameter_description_json: A JSON description of the function
+        parameter, which is provided to the LLM.
+        If no description is provided, the description will be autogenerated.
+        :param model: The OpenAI chat model to use.
         :param data_type: Defines how Functions runtime should treat the
         parameter value.
         :param kwargs: Keyword arguments for specifying additional binding
@@ -1483,7 +1492,7 @@ def decorator():
                 fb.add_trigger(
                     trigger=AssistantSkillTrigger(
                         name=arg_name,
-                        function_description=task_description,
+                        function_description=function_description,
                         data_type=parse_singular_param_to_enum(data_type,
                                                                DataType),
                         **kwargs))
@@ -2813,8 +2822,7 @@ def text_completion_input(self,
         the function, which then returns the result text as the response
         content.
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param prompt: The prompt to generate completions for, encoded as a
         string.
         :param model: the ID of the model to use.
@@ -2867,8 +2875,7 @@ def assistant_create_output(self, arg_name: str,
         The assistantCreate output binding creates a new assistant with a
         specified system prompt.
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param data_type: Defines how Functions runtime should treat the
         parameter value
         :param kwargs: Keyword arguments for specifying additional binding
@@ -2904,8 +2911,7 @@ def assistant_query_input(self,
         The assistantQuery input binding fetches the assistant history and
         passes it to the function.
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param timestamp_utc: the timestamp of the earliest message in the chat
         history to fetch. The timestamp should be in ISO 8601 format - for
         example, 2023-08-01T00:00:00Z.
@@ -2947,8 +2953,7 @@ def assistant_post_input(self, arg_name: str,
         The assistantPost output binding sends a message to the assistant and
         saves the response in its internal state.
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param id: The ID of the assistant to update.
         :param user_message: The user message that user has entered for
         assistant to respond to.
@@ -2992,12 +2997,11 @@ def embeddings_input(self,
             -> Callable[..., Any]:
         """
         The embeddings input decorator creates embeddings which will be used to
-        measure the readiness of text strings.
+        measure the relatedness of text strings.
 
         Ref: https://platform.openai.com/docs/guides/embeddings
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param input: The input source containing the data to generate
         embeddings for.
         :param input_type: The type of the input.
@@ -3048,13 +3052,17 @@ def semantic_search_input(self,
                               **kwargs) \
             -> Callable[..., Any]:
         """
-        The embeddings input decorator creates embeddings which will be used to
-        measure the readiness of text strings.
+        The semantic search feature allows you to import documents into a
+        vector database using an output binding and query the documents in that
+        database using an input binding. For example, you can have a function
+        that imports documents into a vector database and another function that
+        issues queries to OpenAI using content stored in the vector database as
+         context (also known as the Retrieval Augmented Generation, or RAG
+         technique).
 
         Ref: https://platform.openai.com/docs/guides/embeddings
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param connection_name: app setting or environment variable which
         contains a connection string value.
         :param collection: The name of the collection or table to search or
@@ -3064,7 +3072,7 @@ def semantic_search_input(self,
         The default value is "text-embedding-ada-002".
         :param chat_model: The name of the Large Language Model to invoke for
         chat responses. The default value is "gpt-3.5-turbo".
-        :param system_prompt: Optional The system prompt to use for prompting
+        :param system_prompt: Optional. The system prompt to use for prompting
         the large language model.
         :param max_knowledge_count: Optional. The number of knowledge items to
         inject into the SystemPrompt. Default value: 1
@@ -3121,8 +3129,7 @@ def embeddings_store_output(self,
         - Azure Data Explorer
         - Azure Cosmos DB using MongoDB
 
-        :param arg_name: The name of the variable that represents DaprState
-        output object in function code.
+        :param arg_name: The name of binding parameter in the function code.
         :param input: The input to generate embeddings for.
         :param input_type: The type of the input.
         :param connection_name: The name of an app setting or environment

azure/functions/decorators/openai.py

@@ -32,9 +32,15 @@ def get_binding_name() -> str:
     def __init__(self,
                  name: str,
                  function_description: str,
+                 function_name: Optional[str] = None,
+                 parameter_description_json: Optional[str] = None,
+                 model: Optional[OpenAIModels] = OpenAIModels.DefaultChatModel,
                  data_type: Optional[DataType] = None,
                  **kwargs):
         self.function_description = function_description
+        self.function_name = function_name
+        self.parameter_description_json = parameter_description_json
+        self.model = model
         super().__init__(name=name, data_type=data_type)
 
 

tests/decorators/test_openai.py

@@ -13,13 +13,19 @@ class TestOpenAI(unittest.TestCase):
     def test_assistant_skills_trigger_valid_creation(self):
         trigger = AssistantSkillTrigger(name="test",
                                         function_description="description",
+                                        function_name="test_function_name",
+                                        parameter_description_json="test_json",
+                                        model=OpenAIModels.DefaultChatModel,
                                         data_type=DataType.UNDEFINED,
                                         dummy_field="dummy")
         self.assertEqual(trigger.get_binding_name(),
                          "assistantSkillsTrigger")
         self.assertEqual(
             trigger.get_dict_repr(), {"name": "test",
                                       "functionDescription": "description",
+                                      "functionName": "test_function_name",
+                                      "parameterDescriptionJson": "test_json",
+                                      "model": OpenAIModels.DefaultChatModel,
                                       "dataType": DataType.UNDEFINED,
                                       'type': 'assistantSkillsTrigger',
                                       'dummyField': 'dummy',