chat-object 2.0.1

A simple library for creating and manipulating chat and message objects for LLM applications

chat-object

A simple library for creating and managing chat objects and messages for LLM applications.

Installation

From PyPI:

pip install chat-object

From GitHub:

pip install git+https://github.com/fresh-milkshake/chat-object.git

Or from source:

git clone https://github.com/fresh-milkshake/chat-object.git
cd chat-object
pip install -e .

Quick Start

Basic Chat Usage

Create a chat object and add messages to it:

import openai
from chat_object import Chat, Message, Role

client = openai.OpenAI()

chat = Chat(
    Message(Role.System, "You are a helpful assistant"),
    Message(Role.User, "Hello!")
)

response = client.chat.completions.create(
    model="gpt-5-nano",
    messages=chat.as_dict()
)

print(response.choices[0].message.content)

Using the Prompt Class

The Prompt class automatically handles indentation and formatting:

from chat_object import Prompt

# Clean indentation automatically
prompt = Prompt("""
    You are a helpful assistant.
    Please help me with the following task:
    
    def example_function():
        return "hello world"
    
    Explain what this function does.
""")

# Multiple arguments are joined with newlines
prompt = Prompt(
    "You are a helpful assistant.",
    "Please be concise in your responses.",
    "Focus on practical solutions."
)

# String operations work naturally
prompt += "\n\nAdditional context here"

QOL Features for Quick Development

Use convenience functions for faster development:

from chat_object import chat, msg_user, msg_system, msg_assistant, prmt

# Quick chat creation
chat_obj = chat(
    msg_system("You are a helpful assistant."),
    msg_user("Hello!"),
    msg_assistant("Hi there! How can I help you today?")
)

# Quick prompt creation
prompt = prmt("You are a helpful assistant.")

# Convert to dict for API calls
messages = chat_obj.as_dict()

[!TIP] See examples folder for more comprehensive examples.

Features

• Well-tested code: Comprehensive test coverage with doctests throughout the codebase (90% coverage)
• Type safety: Full type hints and enum-based roles
• Backward compatibility: seamless integration with existing APIs like OpenAI, Anthropic, Together, Ollama, etc.
• QOL features: Quick and easy message creation with msg_user, msg_assistant, msg_system, prmt, msgs, chat (Recommended, but not required). Pretty rich example usage of qol features is in examples/openai_use_case.py.

License

This project is licensed under the MIT License. See the LICENSE file for details.