Update agent-apis example

Author: leawn

agent_apis/pyproject.toml

@@ -11,7 +11,7 @@ dependencies = [
     "python-dotenv==1.0.1",
     "openai>=1.61.0",
     "aiohttp>=3.11.12",
-    "restack-ai>=0.0.77",
+    "restack-ai>=0.0.78",
 ]
 
 [project.scripts]

agent_apis/src/workflows/multistep.py

@@ -1,7 +1,7 @@
 from datetime import timedelta
 
 from pydantic import BaseModel, Field
-from restack_ai.workflow import import_functions, log, workflow
+from restack_ai.workflow import import_functions, log, workflow, NonRetryableError
 
 with import_functions():
     from src.functions.llm import FunctionInputParams, llm
@@ -20,20 +20,28 @@ async def run(self, workflow_input: WorkflowInputParams) -> dict:
         user_content = f"Greet this person {workflow_input.name}"
 
         # Step 1 get weather data
-        weather_data = await workflow.step(
-            function=weather, start_to_close_timeout=timedelta(seconds=120)
-        )
-
-        # Step 2 Generate greeting with LLM  based on name and weather data
-
-        llm_message = await workflow.step(
-            function=llm,
-            function_input=FunctionInputParams(
-                system_content=f"You are a personal assitant and have access to weather data {weather_data}. Always greet person with relevant info from weather data",
-                user_content=user_content,
-                model="gpt-4o-mini",
-            ),
-            start_to_close_timeout=timedelta(seconds=120),
-        )
-        log.info("MultistepWorkflow completed", llm_message=llm_message)
-        return {"message": llm_message, "weather": weather_data}
+        try:
+            weather_data = await workflow.step(
+                weather, start_to_close_timeout=timedelta(seconds=120)
+            )
+        except Exception as e:
+            error_message = f"Error during weather: {e}"
+            raise NonRetryableError(error_message) from e
+        else:
+            # Step 2 Generate greeting with LLM  based on name and weather data
+            try:
+                llm_message = await workflow.step(
+                    function=llm,
+                    function_input=FunctionInputParams(
+                        system_content=f"You are a personal assitant and have access to weather data {weather_data}. Always greet person with relevant info from weather data",
+                        user_content=user_content,
+                        model="gpt-4o-mini",
+                    ),
+                    start_to_close_timeout=timedelta(seconds=120),
+                )
+            except Exception as e:
+                error_message = f"Error during llm: {e}"
+                raise NonRetryableError(error_message) from e
+            else:
+                log.info("MultistepWorkflow completed", llm_message=llm_message)
+                return {"message": llm_message, "weather": weather_data}