chore: use ainvoke in samples, improve interrupt models docs

Author: radu-mocanu

docs/interrupt_models.md

@@ -19,10 +19,11 @@ Upon completion of the invoked process, the current agent will automatically res
 
 #### Example:
 ```python
+from uipath.models import InvokeProcess
 process_output = interrupt(InvokeProcess(name="MyProcess", input_arguments={"arg1": "value1"}))
 ```
 
-For a practical implementation of the `InvokeProcess` model, refer to the sample usage in the [planner.py](../../samples/multi-agent-planner-researcher-coder-distributed/src/multi-agent-distributed/planner.py#L184) file. This example demonstrates how to invoke a process with dynamic input arguments, showcasing the integration of the interrupt functionality within a multi-agent system or a system where an agent integrates with RPA processes and API workflows.
+For a practical implementation of the `InvokeProcess` model, refer to the sample usage in the [planner.py](../samples/multi-agent-planner-researcher-coder-distributed/src/multi-agent-distributed/planner.py#L181) file. This example demonstrates how to invoke a process with dynamic input arguments, showcasing the integration of the interrupt functionality within a multi-agent system or a system where an agent integrates with RPA processes and API workflows.
 
 ---
 
@@ -36,6 +37,7 @@ The `WaitJob` model is used to wait for a job completion. Unlike `InvokeProcess`
 
 #### Example:
 ```python
+from uipath.models import WaitJob
 job_output = interrupt(WaitJob(job=my_job_instance))
 ```
 
@@ -57,8 +59,10 @@ For more information on UiPath apps, refer to the [UiPath Apps User Guide](https
 
 #### Example:
 ```python
+from uipath.models import CreateAction
 action_output = interrupt(CreateAction(name="AppName", title="Escalate Issue", data={"key": "value"}, app_version=1, assignee="user@example.com"))
 ```
+For a practical implementation of the `CreateAction` model, refer to the sample usage in the [ticket-classification/main.py](../samples/ticket-classification/main.py#L116) file. This example demonstrates how to create an action with dynamic input.
 
 ---
 
@@ -71,6 +75,7 @@ The `WaitAction` model is used to wait for an action to be handled. This model i
 
 #### Example:
 ```python
+from uipath.models import WaitAction
 action_output = interrupt(WaitAction(action=my_action_instance))
 ```
 

samples/company-research-agent/graph.py

@@ -43,7 +43,7 @@ class GraphOutput(BaseModel):
     response: str
 
 
-def research_node(state: GraphInput) -> GraphOutput:
+async def research_node(state: GraphInput) -> GraphOutput:
     # Format the user message with the company name
     user_message = f"""Please provide a comprehensive analysis and outreach strategy for the company: {state.company_name}. Use the TavilySearchResults tool to gather information. Include detailed research on the company's background, organizational structure, key decision-makers, and a tailored outreach strategy. Format your response using the following section headers:
 
@@ -58,7 +58,7 @@ def research_node(state: GraphInput) -> GraphOutput:
 
     new_state = MessagesState(messages=[{"role": "user", "content": user_message}])
 
-    result = research_agent.invoke(new_state)
+    result = await research_agent.ainvoke(new_state)
 
     return GraphOutput(response=result["messages"][-1].content)
 

samples/multi-agent-planner-researcher-coder-distributed/pyproject.toml

@@ -12,7 +12,7 @@ dependencies = [
     "langchain-anthropic>=0.3.8",
     "langchain-experimental>=0.3.4",
     "tavily-python>=0.5.0",
-    "uipath-langchain==0.0.88"
+    "uipath-langchain==0.0.90"
 ]
 
 [project.optional-dependencies]

samples/multi-agent-planner-researcher-coder-distributed/src/multi-agent-distributed/coder.py

@@ -36,8 +36,8 @@ def python_repl_tool(
 code_agent = create_react_agent(llm, tools=[python_repl_tool])
 
 
-def code_node(state: MessagesState) -> GraphOutput:
-    result = code_agent.invoke(state)
+async def code_node(state: MessagesState) -> GraphOutput:
+    result = await code_agent.ainvoke(state)
     return GraphOutput(answer=result["messages"][-1].content)
 
 

samples/multi-agent-planner-researcher-coder-distributed/src/multi-agent-distributed/planner.py

@@ -70,7 +70,7 @@ def input(state: GraphInput):
 llm = ChatAnthropic(model="claude-3-5-sonnet-latest")
 
 
-def create_plan(state: State) -> Command:
+async def create_plan(state: State) -> Command:
     """Create an execution plan based on the user's question."""
     parser = PydanticOutputParser(pydantic_object=ExecutionPlan)
 
@@ -104,7 +104,7 @@ def create_plan(state: State) -> Command:
         format_instructions=parser.get_format_instructions(),
     )
 
-    plan_response = llm.invoke(formatted_prompt)
+    plan_response = await llm.ainvoke(formatted_prompt)
 
     try:
         plan_output = parser.parse(plan_response.content)

samples/multi-agent-planner-researcher-coder-distributed/src/multi-agent-distributed/researcher.py

@@ -18,8 +18,8 @@ class GraphOutput(BaseModel):
     answer: str
 
 
-def research_node(state: MessagesState) -> GraphOutput:
-    result = research_agent.invoke(state)
+async def research_node(state: MessagesState) -> GraphOutput:
+    result = await research_agent.ainvoke(state)
     return GraphOutput(answer=result["messages"][-1].content)