Skip to content

README.md

Latest commit

 

History

History
71 lines (49 loc) · 2.75 KB

README.md

File metadata and controls

71 lines (49 loc) · 2.75 KB
Raw

Basic Multi-Agent Example (AutoGen)

This folder contains a minimal Jupyter Notebook (main.ipynb) that demonstrates a very small multi‑agent workflow using AutoGen AgentChat. Two assistant agents collaborate in a round‑robin group chat to produce and refine a short joke about a chosen topic.

What It Does

  1. Loads a model configuration via load_model_config().
  2. Defines two agents:
    • Author – writes a SHORT joke about the topic.
    • SmartassEditor – analyzes why it is (or isn’t) funny and makes it funnier without making it longer.
  3. Runs a RoundRobinGroupChat, allowing each agent a single turn to contribute in sequence.
  4. Prints the message sequence plus timing & stop info.

Example Output

MESSAGE 0 [user]:
Invent a short joke about: Animals

----------------------------------------------------------------
MESSAGE 1 [Author]:
Why don't animals play poker in the jungle? Too many cheetahs.

----------------------------------------------------------------
MESSAGE 2 [SmartassEditor]:
Why it's funny: the joke is a classic pun — "cheetahs" sounds like "cheaters" — combined with the absurd image of animals playing poker. The humor comes from the wordplay and the surprise of the literal animal answer; it's a dad-joke style gag, so it's groan-inducing rather than sophisticated.

Funnier version (shorter and punchier):
Why don't jungle animals play poker? Too many cheetahs.

Quick Start

  1. Activate your virtual environment in the project root.

  2. Install dependencies (example with uv):

    uv sync

    Or with pip (editable install):

    pip install -e .

  3. Configure model access

    OpenAI or Azure OpenAI API key require one of OPENAI_API_KEY or AZURE_OPENAI_API_KEY env variables set (loaded via .env if present). Alternatively, a local ollama server can be used with a suitable model (changed in load_model_config()).

  4. Open basic_example/main.ipynb in VS Code.

  5. Run cells top to bottom; final cells show each message and stop reason.

Customizing

  • Change the topic in topic = "Animals".
  • Add or modify roles in the agentConfig dict.
  • Uncomment adjust_properties_for_fixed_response(...) for a deterministic test.
  • Use a different model by modifying the load_model_config() function in main.py.

Key Functions

  • create_chat_completion_client(model) – constructs the model client wrapper.
  • AssistantAgent – defines an autonomous role with a system prompt.
  • RoundRobinGroupChat – simple orchestrator cycling agents.
  • invent_joke(topic) – async coroutine that runs the interaction.

Brief by design—see the notebook and tools.py for full details. The notebook closes the model_client explicitly (await model_client.close()).