MinAgent

Agent is a wrapper on top of OpenAI chat completion and assistant API. It handles tool_call, file_search within the class

Features

Implemented

• Chat Completion API
• Assistant API

And tools

• Function call (Parallel execution)
• File search

Usage

pip install minagent

Create an Agent

from openai import OpenAI
from minagent import Agent

openai_client = OpenAI(
    api_key="00000"
)

def current_utc_datetime(**args) -> str:
    from datetime import datetime, timezone
    return datetime.now(timezone.utc).isoformat()

clock = Agent(
    openai_client,
    name='clock',
    model='gpt-4o',
    instructions=['You are a time aware assistant.',],

    # optional: only add below if using assistant api
    openai_asst_id='asst_0000'

    # optional: tool for file_search, and function
    tools=[
        {
            'type': 'file_search',
        }, {
            'type': 'function',
            'function': {"name": 'current_utc_datetime'},
        },
    ],
    tool_functions={'current_utc_datetime': current_utc_datetime},
    tool_resources={'file_search': {'vector_store_ids': ['vs_0000']}},
)

Use of Chat completion API

response = clock.chat_completion(
    user='john',
    messages=[
        { 'role': 'user', 'content': 'what time is it now?' },
    ],
)

Use of Assistant API

thread = openai_client.beta.threads.create()
response = clock.ask_assistant(
    user='john',
    asst_thread_id=thread.id,
    message=[
        { 'role': 'user', 'content': 'what time is it now?' },
    ]
)

Future

• Support for Response API
• Add test cases
• Release pipeline