agent-framework-chatkit

What it is and what it does

This package is a bridge layer that lets you run a Microsoft Agent Framework agent as the backend for an OpenAI ChatKit chat interface. It handles the translation between the two systems: converting ChatKit thread messages into the format Agent Framework expects, and converting Agent Framework's streamed responses back into ChatKit events that the frontend can display.

The package provides three main helpers: `simple_to_agent_input` to quickly convert a ChatKit thread into Agent Framework messages, `stream_agent_response` to stream agent outputs back to ChatKit, and `ThreadItemConverter` as an extensible class for custom message transformations. It depends on agent-framework-core and openai-chatkit. The frontend UI itself requires internet access to OpenAI's CDN and is not self-hostable, though the backend components are fully open source.

Use it for:

• Build a chat application where an Agent Framework agent with custom tools responds to user messages through a ChatKit UI.
• Stream long-running agent operations (tool calls, reasoning steps) to a ChatKit frontend in real time.
• Convert existing ChatKit thread history into Agent Framework message format to maintain conversation context across agent runs.
• Extend ThreadItemConverter to handle domain-specific message types or metadata from ChatKit threads.
• Deploy an agentic backend that integrates with OpenAI's ChatKit platform without building custom frontend code.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Bridges Microsoft Agent Framework agents with OpenAI ChatKit, converting between Agent Framework message formats and ChatKit thread events for integrated agentic chat applications.

Yes, if you are building an Agent Framework agent and want to expose it through ChatKit's UI. The package is actively maintained, has low install friction, and provides the exact integration layer needed. However, be aware that ChatKit's frontend requires internet connectivity to OpenAI's CDN—this is a hard blocker for air-gapped or highly regulated environments. The package is still in beta; verify that the threading and streaming behavior meets your production requirements before committing.

Install

agent-framework-chatkit on PyPI

pip install agent-framework-chatkit

uv add agent-framework-chatkit

poetry add agent-framework-chatkit

Installing agent-framework-chatkit

Before you install

Low friction install with two direct dependencies. Package is in beta (version 1.0.0b260730) but actively maintained—released 15 days ago with recent commits. Supports Python 3.10 through 3.13.

License in practice

MIT license (permissive). No restrictions on commercial or private use.

Quickstart

pip install agent-framework-chatkit --pre

from agent_framework_chatkit import simple_to_agent_input, stream_agent_response

# Convert ChatKit thread items to Agent Framework messages
agent_messages = await simple_to_agent_input(thread_items)

# Stream agent responses back to ChatKit events
async for event in stream_agent_response(response_stream, thread_id):
    yield event

Requires Python 3.10+. ChatKit frontend requires internet connectivity to OpenAI's CDN (cdn.platform.openai.com); not suitable for air-gapped environments.

Verify before relying

• Whether ThreadItemConverter's extensibility covers all custom message types needed in production deployments
• Performance characteristics when streaming large conversation histories through simple_to_agent_input
• Compatibility guarantees between agent-framework-core and openai-chatkit versions

Package facts

Evidence: agent_framework_chatkit-1.0.0b260730-py3-none-any.whl

Tags

More Application Frameworks packages

Further reading