Merge pull request #177 from PrefectHQ/tasks-and-agents

Add tasks and agents guide

docs/examples/call-routing.mdx

@@ -0,0 +1,73 @@
+---
+title: Customer Call Routing
+---
+
+In this example, two agents interact in a call routing scenario. One agent plays the role of a customer calling into a call center, while the other agent is a trainee customer service representative. The trainee must listen to the customer's story and route them to the correct department based on the information provided.
+
+
+```python
+import random
+from enum import Enum
+import controlflow as cf
+
+
+class Department(Enum):
+    SALES = "sales"
+    SUPPORT = "support"
+    BILLING = "billing"
+    TECHNICAL = "technical"
+    RETURNS = "returns"
+
+
+
+@cf.flow
+def routing_flow():
+    department = random.choice(list(Department))
+
+    # create an agent to be our "customer"
+    customer = cf.Agent(
+        "Customer",
+        instructions=f"""
+            You are training customer reps by pretending to be a customer
+            calling into a call center. You need to be routed to the
+            {department} department. Come up with a good backstory.""",
+    )
+
+    trainee = cf.Agent(
+        "Trainee",
+        instructions="""
+            You are a trainee customer service representative. You need to
+            listen to the customer's story and route them to the correct
+            department. Note that the customer is another agent training you.""",
+    )
+
+    task = cf.Task(
+        """
+        In this task, the customer agent and the trainee agent will speak to
+        each other. The customer is trying to be routed to the correct
+        department. The trainee will listen and ask questions to determine the
+        correct department. 
+        """,
+        instructions="""
+        Only the trainee can mark the task successful by routing the customer to
+        the correct department. The trainee must begin the conversation by
+        greeting the customer. Agents speak to each other by posting messages
+        directly to the thread. Do not use the `end_turn` tool or try to talk 
+        to a user.
+        """,
+        agents=[trainee, customer],
+        result_type=Department,
+    )
+
+    routed_dapartment = task.run()
+    if routed_dapartment == department:
+        print("Success! The customer was routed to the correct department.")
+    else:
+        print(
+            "Failed. The customer was routed to the wrong department. "
+            f"The correct department was {department}."
+        )
+
+if __name__ == "__main__":
+    routing_flow()
+````

docs/examples/library.mdx

@@ -3,9 +3,12 @@ title: Library
 ---
 
 <CardGroup cols={2}>
-  <Card title="Rock, Paper, Scissors" icon="hand-scissors" href="/examples/rock_paper_scissors">
+  <Card title="Rock, Paper, Scissors" icon="hand-scissors" href="/examples/rock-paper-scissors">
     Play a game of rock, paper, scissors against an AI - without letting it cheat.
   </Card>
+  <Card title="Customer Call Routing" icon="phone" href="/examples/call-routing">
+    Two agents cooperate to route customer calls to the correct department.
+  </Card>
   <Card title="Don't Panic." icon="paper-plane">
     More examples are coming soon.
   </Card>

docs/guides/agentic-loop.mdx

@@ -1,6 +1,5 @@
 ---
-title: Working with the Agentic Loop
-sidebarTitle: "Agentic Loops"
+title: Managing the Agentic Loop
 ---
 
 The **agentic loop** is a fundamental concept in agentic workflows, representing the iterative process of invoking AI agents to make progress towards a goal. It is at the heart of every agentic workflow because agents almost always require multiple invocations to complete complex tasks.