Lab L1.2: Hello World Agent

🎯 Assignment: Accept this lab on GitHub Classroom
You’ll get your own repository with starter code, instructions, and automatic grading.


Duration	45 minutes
Prerequisites	Previous module completed

Objectives

Create a minimal working agent
Configure basic prompt and voice
Test with swaig-test CLI
Understand generated SWML

How to Complete This Lab

Accept the Assignment — Click the GitHub Classroom link above
Clone Your Repo — git clone <your-repo-url>
Read the README — Your repo has detailed requirements and grading criteria
Write Your Code — Implement the solution in solution/agent.py
Test Locally — Use swaig-test to verify your agent works
Push to Submit — git push triggers auto-grading

Key Concepts

The following exercises walk through the concepts you’ll need. Your GitHub Classroom repo README has the specific requirements for grading.

Exercise 1: Minimal Agent

Create the simplest possible agent:

Create hello_agent.py:

#!/usr/bin/env python3
"""Minimal Hello World agent."""

from signalwire_agents import AgentBase

# Create agent with just a name
agent = AgentBase(name="hello-agent")

if __name__ == "__main__":
    agent.run()

Test with swaig-test:

swaig-test hello_agent.py --dump-swml

Questions to Answer:

What does the SWML contain?
Is there a prompt defined?
What language/voice is configured?

Exercise 2: Add Prompt

Update the agent with a basic prompt:

Update hello_agent.py:

#!/usr/bin/env python3
"""Hello World agent with prompt."""

from signalwire_agents import AgentBase

agent = AgentBase(name="hello-agent")

# Add a prompt
agent.prompt_add_section(
    "Role",
    "You are a friendly assistant. When someone calls, "
    "greet them warmly and ask how you can help today."
)

if __name__ == "__main__":
    agent.run()

Test again:

swaig-test hello_agent.py --dump-swml

Questions to Answer:

Where does your prompt appear in the SWML?
How is the section title used?

Exercise 3: Add Voice Configuration

Add language and voice:

Update hello_agent.py:

#!/usr/bin/env python3
"""Hello World agent with prompt and voice."""

from signalwire_agents import AgentBase

agent = AgentBase(name="hello-agent")

agent.prompt_add_section(
    "Role",
    "You are a friendly assistant. When someone calls, "
    "greet them warmly and ask how you can help today."
)

# Add voice configuration
agent.add_language("English", "en-US", "rime.spore")

if __name__ == "__main__":
    agent.run()

Test:

swaig-test hello_agent.py --dump-swml

Questions to Answer:

What does the languages array contain?
What fields are in each language object?

Exercise 4: Examine SWML Structure

Save and analyze the SWML:

swaig-test hello_agent.py --dump-swml --raw > hello_swml.json

Open hello_swml.json in your editor.

Identify these elements:

Draw the hierarchy:

{
  version
  sections
    main
      ai
        prompt
          text
        languages
          [0]
            name
            code
            voice
}

Exercise 5: Class-Based Agent

Convert to class-based pattern:

Create hello_class_agent.py:

#!/usr/bin/env python3
"""Hello World agent using class pattern."""

from signalwire_agents import AgentBase


class HelloAgent(AgentBase):
    """A friendly greeting agent."""

    def __init__(self):
        super().__init__(name="hello-class-agent")
        self._configure_prompt()
        self._configure_voice()

    def _configure_prompt(self):
        """Set up the agent's prompt."""
        self.prompt_add_section(
            "Role",
            "You are a friendly assistant. When someone calls, "
            "greet them warmly and ask how you can help today."
        )

    def _configure_voice(self):
        """Set up voice and language."""
        self.add_language("English", "en-US", "rime.spore")


if __name__ == "__main__":
    agent = HelloAgent()
    agent.run()

Test:

swaig-test hello_class_agent.py --dump-swml

Questions to Answer:

Is the output different from the functional version?
What are the benefits of the class-based pattern?

Exercise 6: Run Locally

Start your agent server:

python hello_agent.py

Expected Output:

INFO:     Started server process [XXXXX]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:3000

In another terminal, test with curl:

curl http://localhost:3000/

You should get the SWML JSON response.

Stop the server: Press Ctrl+C

Exercise 7: Change Port

Run on a different port:

PORT=5000 python hello_agent.py

Test:

curl http://localhost:5000/

Challenge Exercise (Optional)

Create an agent with:

Custom name: “greeter-bot”
Custom route: “/greet”
Prompt with multiple sections
Different voice (try rime.marsh)

Verify:

swaig-test your_agent.py --dump-swml

Deliverables

Submit or show instructor:

hello_agent.py - functional pattern
hello_class_agent.py - class pattern
hello_swml.json - generated SWML

Review Questions

What is the minimum required to create an agent?
What does agent.run() do?
Why use if __name__ == "__main__"?
What’s the difference between functional and class-based patterns?
Where does the prompt appear in SWML?

Summary

You have successfully:

Created a minimal agent
Added prompt configuration
Added voice configuration
Examined generated SWML
Ran the agent locally

Next: Lab 1.3 - SWML Inspection

Complete Agent Code

Click to reveal complete solution

#!/usr/bin/env python3
"""Hello World agent with prompt and voice - Functional Pattern.

Lab 1.2 Deliverable: Demonstrates the functional approach to creating
a SignalWire AI agent with basic prompt and voice configuration.
"""

from signalwire_agents import AgentBase

agent = AgentBase(name="hello-agent")

# Add prompt configuration
agent.prompt_add_section(
    "Role",
    "You are a friendly assistant. When someone calls, "
    "greet them warmly and ask how you can help today."
)

# Add voice configuration
agent.add_language("English", "en-US", "rime.spore")

if __name__ == "__main__":
    agent.run()