Add Microsoft Agent Framework Python integration support #645
+1,738
−1
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
moonbox3
requested review from
NathanTarbert,
ataibarkai,
maxkorp,
mme,
ranst91 and
tylerslaton
as code owners
moonbox3
commented
Nov 6, 2025
| { | ||
| "name": "tool_based_generative_ui.py", | ||
| "content": "\"\"\"\nAn example demonstrating tool-based generative UI.\n\"\"\"\n\nfrom crewai.flow.flow import Flow, start\nfrom litellm import completion\nfrom ..sdk import copilotkit_stream, CopilotKitState\n\n\n# This tool generates a haiku on the server.\n# The tool call will be streamed to the frontend as it is being generated.\nGENERATE_HAIKU_TOOL = {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"generate_haiku\",\n \"description\": \"Generate a haiku in Japanese and its English translation\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"japanese\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"An array of three lines of the haiku in Japanese\"\n },\n \"english\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"An array of three lines of the haiku in English\"\n },\n \"image_names\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"Names of 3 relevant images from the provided list\"\n }\n },\n \"required\": [\"japanese\", \"english\", \"image_names\"]\n }\n }\n}\n\n\nclass ToolBasedGenerativeUIFlow(Flow[CopilotKitState]):\n \"\"\"\n A flow that demonstrates tool-based generative UI.\n \"\"\"\n\n @start()\n async def chat(self):\n \"\"\"\n The main function handling chat and tool calls.\n \"\"\"\n system_prompt = \"You assist the user in generating a haiku. When generating a haiku using the 'generate_haiku' tool, you MUST also select exactly 3 image filenames from the following list that are most relevant to the haiku's content or theme. Return the filenames in the 'image_names' parameter. Dont provide the relavent image names in your final response to the user. \"\n\n\n # 1. Run the model and stream the response\n # Note: In order to stream the response, wrap the completion call in\n # copilotkit_stream and set stream=True.\n response = await copilotkit_stream(\n completion(\n\n # 1.1 Specify the model to use\n model=\"openai/gpt-4o\",\n messages=[\n {\n \"role\": \"system\", \n \"content\": system_prompt\n },\n *self.state.messages\n ],\n\n # 1.2 Bind the available tools to the model\n tools=[ GENERATE_HAIKU_TOOL ],\n\n # 1.3 Disable parallel tool calls to avoid race conditions,\n # enable this for faster performance if you want to manage\n # the complexity of running tool calls in parallel.\n parallel_tool_calls=False,\n stream=True\n )\n )\n message = response.choices[0].message\n\n # 2. Append the message to the messages in state\n self.state.messages.append(message)\n\n # 3. If there are tool calls, append a tool message to the messages in state\n if message.tool_calls:\n self.state.messages.append(\n {\n \"tool_call_id\": message.tool_calls[0].id,\n \"role\": \"tool\",\n \"content\": \"Haiku generated.\"\n }\n )\n", | ||
| "content": "\"\"\"\nAn example demonstrating tool-based generative UI.\n\"\"\"\n\nfrom crewai.flow.flow import Flow, start\nfrom litellm import completion\nfrom ag_ui.core import MessagesSnapshotEvent, EventType\nfrom ..sdk import copilotkit_stream, CopilotKitState\n\n\n# This tool generates a haiku on the server.\n# The tool call will be streamed to the frontend as it is being generated.\nGENERATE_HAIKU_TOOL = {\n \"type\": \"function\",\n \"function\": {\n \"name\": \"generate_haiku\",\n \"description\": \"Generate a haiku in Japanese and its English translation\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"japanese\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"An array of three lines of the haiku in Japanese\"\n },\n \"english\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"An array of three lines of the haiku in English\"\n },\n \"image_names\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"Names of 3 relevant images from the provided list\"\n }\n },\n \"required\": [\"japanese\", \"english\", \"image_names\"]\n }\n }\n}\n\n\nclass ToolBasedGenerativeUIFlow(Flow[CopilotKitState]):\n \"\"\"\n A flow that demonstrates tool-based generative UI.\n \"\"\"\n\n @start()\n async def chat(self):\n \"\"\"\n The main function handling chat and tool calls.\n \"\"\"\n system_prompt = \"You assist the user in generating a haiku. When generating a haiku using the 'generate_haiku' tool, you MUST also select exactly 3 image filenames from the following list that are most relevant to the haiku's content or theme. Return the filenames in the 'image_names' parameter. Dont provide the relavent image names in your final response to the user. \"\n\n\n # 1. Run the model and stream the response\n # Note: In order to stream the response, wrap the completion call in\n # copilotkit_stream and set stream=True.\n response = await copilotkit_stream(\n completion(\n\n # 1.1 Specify the model to use\n model=\"openai/gpt-4o\",\n messages=[\n {\n \"role\": \"system\", \n \"content\": system_prompt\n },\n *self.state.messages\n ],\n\n # 1.2 Bind the available tools to the model\n tools=[ GENERATE_HAIKU_TOOL ],\n\n # 1.3 Disable parallel tool calls to avoid race conditions,\n # enable this for faster performance if you want to manage\n # the complexity of running tool calls in parallel.\n parallel_tool_calls=False,\n stream=True\n )\n )\n message = response.choices[0].message\n\n # 2. Append the message to the messages in state\n self.state.messages.append(message)\n\n # 3. If there are tool calls, append a tool message to the messages in state\n if message.tool_calls:\n self.state.messages.append(\n {\n \"tool_call_id\": message.tool_calls[0].id,\n \"role\": \"tool\",\n \"content\": \"Haiku generated.\"\n }\n )\n \n # 4. Emit MessagesSnapshotEvent to notify frontend about tool result\n yield MessagesSnapshotEvent(\n type=EventType.MESSAGES_SNAPSHOT,\n messages=self.state.messages\n )\n", |
There was a problem hiding this comment.
This was auto-generated when I started the Dojo frontend. Should I keep or revert?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add Microsoft Agent Framework Python Integration for Dojo
This PR adds a complete Python integration for the Microsoft Agent Framework with the AG-UI protocol, demonstrating all 7 Dojo features.
Changes
New Integration (
integrations/microsoft-agent-framework/python/examples/)agent-framework-ag-uipackagepyproject.toml,.env.example, and documentationKey Features
agent-framework-ag-ui==1.0.0b251106.post1from PyPI (includes examples package)agent_framework_ag_ui_examples.agentsmoduleDependencies
agent-framework-ag-ui>=1.0.0b251106.post1agent-framework-core>=1.0.0b251105python-dotenv>=1.0.0Address (Python portion of) Add AG-UI Support for Microsoft Agent Framework #487