Multi-Agent Systems

Hands-on lab with starter code you can copy + run, and guided steps to produce the artifact for this module.

Setup

Open a fresh terminal in a new folder, e.g. mkdir lab-5 && cd lab-5.
Create a Python 3.11 venv: python -m venv .venv && source .venv/bin/activate.
Install the dependencies noted on each starter snippet (the small pip install … line above the code).
Copy each starter into a file with the suggested name, then run it and capture the output.

Starter code

01 · Two-agent debate with a judge

save as: 05_01_lab.py

python

from openai import OpenAI
c = OpenAI()

def speak(role, history):
    return c.chat.completions.create(
        model="gpt-5-mini",
        messages=[{"role":"system","content":f"You are {role}."},
                  {"role":"user","content":history}],
    ).choices[0].message.content

def debate(topic, rounds=3):
    history = f"Topic: {topic}\n"
    for _ in range(rounds):
        history += "\nPRO: " + speak("a debater arguing FOR", history)
        history += "\nCON: " + speak("a debater arguing AGAINST", history)
    verdict = speak("an impartial judge giving a 1-paragraph verdict", history)
    return verdict

print(debate("Should agents have persistent memory by default?"))

Run with: python 05_01_lab.py

Guided steps

Work through each step in order. Stop when your output matches the expected artifact — then push your work to your lab folder.

1
Debate-and-judge
Do — Build PRO/CON debaters and an impartial judge. Run on 5 contested questions.
Expected artifact — debate.md with full transcripts and a 1-paragraph verdict for each.
Run free:
2
Specialist team
Do — Compose 3 agents (Researcher, Coder, Critic). Have them produce a 50-line Python utility.
Expected artifact — util.py file plus a transcript showing each agent's contribution.
Run free:
3
Coordination failure log
Do — Force 5 cases where agents loop or contradict. Document patterns and your fix.
Expected artifact — coord_failures.md with case, symptom, root cause, fix.
Run free:

Authoritative sources

Independent primary sources used to check and corroborate the guidance on this page.

NIST National Initiative for Cybersecurity Education (NICE)
US national framework for cybersecurity work roles and competencies.
Financial Action Task Force (FATF)
Global AML/CFT standards referenced by financial-crime curricula.
FBI Internet Crime Complaint Center — Annual Reports
Primary loss and typology data used in fraud-investigation training.
US Federal Trade Commission — Consumer Sentinel Data Book
Official complaint statistics for teaching fraud trend analysis.