Providers

AutoGen

Markdown

The AutoGen provider transforms Devcaster tools into AutoGen's FunctionTool format for use with AutoGen agents.

Install

pip install devcaster devcaster_autogen autogen-agentchat

Configure API Keys

Set DEVCASTER_API_KEY with your API key from Settings and OPENAI_API_KEY with your OpenAI API key.

.env
DEVCASTER_API_KEY=xxxxxxxxx
OPENAI_API_KEY=xxxxxxxxx

Create session and run

from autogen import AssistantAgent, UserProxyAgent
from devcaster import Devcaster
from devcaster_autogen import AutogenProvider

devcaster = Devcaster(provider=AutogenProvider())

# Create a session for your user
session = devcaster.create(user_id="user_123")
tools = session.tools()

chatbot = AssistantAgent(
    "chatbot",
    system_message="Reply TERMINATE when the task is done or when user's content is empty",
    llm_config={"config_list": [{"model": "gpt-5.2"}]},
)

user_proxy = UserProxyAgent(
    "user_proxy",
    is_termination_msg=lambda msg: "TERMINATE" in (msg.get("content", "") or ""),
    human_input_mode="NEVER",
    code_execution_config={"use_docker": False},
)

# Register tools with both agents
devcaster.provider.register_tools(caller=chatbot, executor=user_proxy, tools=tools)

response = user_proxy.initiate_chat(
    chatbot,
    message="Send an email to john@example.com with the subject 'Hello' and body 'Hello from Devcaster!'",
)

print(response.chat_history)