Capture file IDs from code interpreter in streaming responses (#2741)

This commit is contained in:
Evan Mattson
2025-12-11 16:30:19 +09:00
committed by GitHub
Unverified
parent 4c6a5d4aa1
commit 3481914981
10 changed files with 785 additions and 13 deletions
@@ -14,6 +14,7 @@ This folder contains examples demonstrating different ways to create and use age
| [`azure_ai_with_bing_custom_search.py`](azure_ai_with_bing_custom_search.py) | Shows how to use Bing Custom Search with Azure AI agents to search custom search instances and provide responses with relevant results. Requires a Bing Custom Search connection and instance configured in your Azure AI project. |
| [`azure_ai_with_browser_automation.py`](azure_ai_with_browser_automation.py) | Shows how to use Browser Automation with Azure AI agents to perform automated web browsing tasks and provide responses based on web interactions. Requires a Browser Automation connection configured in your Azure AI project. |
| [`azure_ai_with_code_interpreter.py`](azure_ai_with_code_interpreter.py) | Shows how to use the `HostedCodeInterpreterTool` with Azure AI agents to write and execute Python code for mathematical problem solving and data analysis. |
| [`azure_ai_with_code_interpreter_file_generation.py`](azure_ai_with_code_interpreter_file_generation.py) | Shows how to retrieve file IDs from code interpreter generated files using both streaming and non-streaming approaches. |
| [`azure_ai_with_existing_agent.py`](azure_ai_with_existing_agent.py) | Shows how to work with a pre-existing agent by providing the agent name and version to the Azure AI client. Demonstrates agent reuse patterns for production scenarios. |
| [`azure_ai_with_existing_conversation.py`](azure_ai_with_existing_conversation.py) | Demonstrates how to use an existing conversation created on the service side with Azure AI agents. Shows two approaches: specifying conversation ID at the client level and using AgentThread with an existing conversation ID. |
| [`azure_ai_with_application_endpoint.py`](azure_ai_with_application_endpoint.py) | Demonstrates calling the Azure AI application-scoped endpoint. |
@@ -0,0 +1,111 @@
# Copyright (c) Microsoft. All rights reserved.
import asyncio
from agent_framework import (
CitationAnnotation,
HostedCodeInterpreterTool,
HostedFileContent,
TextContent,
)
from agent_framework._agents import AgentRunResponseUpdate
from agent_framework.azure import AzureAIClient
from azure.identity.aio import AzureCliCredential
"""
Azure AI V2 Code Interpreter File Generation Sample
This sample demonstrates how the V2 AzureAIClient handles file annotations
when code interpreter generates text files. It shows both non-streaming
and streaming approaches to verify file ID extraction.
"""
QUERY = (
"Write a simple Python script that creates a text file called 'sample.txt' containing "
"'Hello from the code interpreter!' and save it to disk."
)
async def test_non_streaming() -> None:
"""Test non-streaming response - should have annotations on TextContent."""
print("=== Testing Non-Streaming Response ===")
async with (
AzureCliCredential() as credential,
AzureAIClient(credential=credential).create_agent(
name="V2CodeInterpreterFileAgent",
instructions="You are a helpful assistant that can write and execute Python code to create files.",
tools=HostedCodeInterpreterTool(),
) as agent,
):
print(f"User: {QUERY}\n")
result = await agent.run(QUERY)
print(f"Agent: {result.text}\n")
# Check for annotations in the response
annotations_found: list[str] = []
# AgentRunResponse has messages property, which contains ChatMessage objects
for message in result.messages:
for content in message.contents:
if isinstance(content, TextContent) and content.annotations:
for annotation in content.annotations:
if isinstance(annotation, CitationAnnotation) and annotation.file_id:
annotations_found.append(annotation.file_id)
print(f"Found file annotation: file_id={annotation.file_id}")
if annotations_found:
print(f"SUCCESS: Found {len(annotations_found)} file annotation(s)")
else:
print("WARNING: No file annotations found in non-streaming response")
async def test_streaming() -> None:
"""Test streaming response - check if file content is captured via HostedFileContent."""
print("\n=== Testing Streaming Response ===")
async with (
AzureCliCredential() as credential,
AzureAIClient(credential=credential).create_agent(
name="V2CodeInterpreterFileAgentStreaming",
instructions="You are a helpful assistant that can write and execute Python code to create files.",
tools=HostedCodeInterpreterTool(),
) as agent,
):
print(f"User: {QUERY}\n")
annotations_found: list[str] = []
text_chunks: list[str] = []
file_ids_found: list[str] = []
async for update in agent.run_stream(QUERY):
if isinstance(update, AgentRunResponseUpdate):
for content in update.contents:
if isinstance(content, TextContent):
if content.text:
text_chunks.append(content.text)
if content.annotations:
for annotation in content.annotations:
if isinstance(annotation, CitationAnnotation) and annotation.file_id:
annotations_found.append(annotation.file_id)
print(f"Found streaming annotation: file_id={annotation.file_id}")
elif isinstance(content, HostedFileContent):
file_ids_found.append(content.file_id)
print(f"Found streaming HostedFileContent: file_id={content.file_id}")
print(f"\nAgent response: {''.join(text_chunks)[:200]}...")
if annotations_found or file_ids_found:
total = len(annotations_found) + len(file_ids_found)
print(f"SUCCESS: Found {total} file reference(s) in streaming")
else:
print("WARNING: No file annotations found in streaming response")
async def main() -> None:
print("AzureAIClient Code Interpreter File Generation Test\n")
await test_non_streaming()
await test_streaming()
if __name__ == "__main__":
asyncio.run(main())
@@ -9,6 +9,7 @@ This folder contains examples demonstrating different ways to create and use age
| [`azure_ai_basic.py`](azure_ai_basic.py) | The simplest way to create an agent using `ChatAgent` with `AzureAIAgentClient`. It automatically handles all configuration using environment variables. |
| [`azure_ai_with_bing_custom_search.py`](azure_ai_with_bing_custom_search.py) | Shows how to use Bing Custom Search with Azure AI agents to find real-time information from the web using custom search configurations. Demonstrates how to set up and use HostedWebSearchTool with custom search instances. |
| [`azure_ai_with_bing_grounding.py`](azure_ai_with_bing_grounding.py) | Shows how to use Bing Grounding search with Azure AI agents to find real-time information from the web. Demonstrates web search capabilities with proper source citations and comprehensive error handling. |
| [`azure_ai_with_code_interpreter_file_generation.py`](azure_ai_with_code_interpreter_file_generation.py) | Shows how to retrieve file IDs from code interpreter generated files using both streaming and non-streaming approaches. |
| [`azure_ai_with_code_interpreter.py`](azure_ai_with_code_interpreter.py) | Shows how to use the HostedCodeInterpreterTool with Azure AI agents to write and execute Python code. Includes helper methods for accessing code interpreter data from response chunks. |
| [`azure_ai_with_existing_agent.py`](azure_ai_with_existing_agent.py) | Shows how to work with a pre-existing agent by providing the agent ID to the Azure AI chat client. This example also demonstrates proper cleanup of manually created agents. |
| [`azure_ai_with_existing_thread.py`](azure_ai_with_existing_thread.py) | Shows how to work with a pre-existing thread by providing the thread ID to the Azure AI chat client. This example also demonstrates proper cleanup of manually created threads. |
@@ -0,0 +1,102 @@
# Copyright (c) Microsoft. All rights reserved.
import asyncio
from agent_framework import AgentRunResponseUpdate, ChatAgent, HostedCodeInterpreterTool, HostedFileContent
from agent_framework.azure import AzureAIAgentClient
from azure.identity.aio import AzureCliCredential
"""
Azure AI Agent Code Interpreter File Generation Example
This sample demonstrates using HostedCodeInterpreterTool with AzureAIAgentClient
to generate a text file and then retrieve it.
The test flow:
1. Create an agent with code interpreter tool
2. Ask the agent to generate a txt file using Python code
3. Capture the file_id from HostedFileContent in the response
4. Retrieve the file using the agents_client.files API
"""
async def main() -> None:
"""Test file generation and retrieval with code interpreter."""
async with AzureCliCredential() as credential:
client = AzureAIAgentClient(credential=credential)
try:
async with ChatAgent(
chat_client=client,
instructions=(
"You are a Python code execution assistant. "
"ALWAYS use the code interpreter tool to execute Python code when asked to create files. "
"Write actual Python code to create files, do not just describe what you would do."
),
tools=[HostedCodeInterpreterTool()],
) as agent:
# Be very explicit about wanting code execution and a download link
query = (
"Use the code interpreter to execute this Python code and then provide me "
"with a download link for the generated file:\n"
"```python\n"
"with open('/mnt/data/sample.txt', 'w') as f:\n"
" f.write('Hello, World! This is a test file.')\n"
"'/mnt/data/sample.txt'\n" # Return the path so it becomes downloadable
"```"
)
print(f"User: {query}\n")
print("=" * 60)
# Collect file_ids from the response
file_ids: list[str] = []
async for chunk in agent.run_stream(query):
if not isinstance(chunk, AgentRunResponseUpdate):
continue
for content in chunk.contents:
if content.type == "text":
print(content.text, end="", flush=True)
elif content.type == "hosted_file":
if isinstance(content, HostedFileContent):
file_ids.append(content.file_id)
print(f"\n[File generated: {content.file_id}]")
print("\n" + "=" * 60)
# Attempt to retrieve discovered files
if file_ids:
print(f"\nAttempting to retrieve {len(file_ids)} file(s):")
for file_id in file_ids:
try:
file_info = await client.agents_client.files.get(file_id)
print(f" File {file_id}: Retrieved successfully")
print(f" Filename: {file_info.filename}")
print(f" Purpose: {file_info.purpose}")
print(f" Bytes: {file_info.bytes}")
except Exception as e:
print(f" File {file_id}: FAILED to retrieve - {e}")
else:
print("No file IDs were captured from the response.")
# List all files to see if any exist
print("\nListing all files in the agent service:")
try:
files_list = await client.agents_client.files.list()
count = 0
for file_info in files_list.data:
count += 1
print(f" - {file_info.id}: {file_info.filename} ({file_info.purpose})")
if count == 0:
print(" No files found.")
except Exception as e:
print(f" Failed to list files: {e}")
finally:
await client.close()
if __name__ == "__main__":
asyncio.run(main())
@@ -0,0 +1,241 @@
# Copyright (c) Microsoft. All rights reserved.
"""
Handoff Workflow with Code Interpreter File Generation Sample
This sample demonstrates retrieving file IDs from code interpreter output
in a handoff workflow context. A triage agent routes to a code specialist
that generates a text file, and we verify the file_id is captured correctly
from the streaming AgentRunUpdateEvent events.
Verifies GitHub issue #2718: files generated by code interpreter in
HandoffBuilder workflows can be properly retrieved.
Toggle USE_V2_CLIENT to switch between:
- V1: AzureAIAgentClient (azure-ai-agents SDK)
- V2: AzureAIClient (azure-ai-projects 2.x with Responses API)
IMPORTANT: When using V2 AzureAIClient with HandoffBuilder, each agent must
have its own client instance. The V2 client binds to a single server-side
agent name, so sharing a client between agents causes routing issues.
Prerequisites:
- `az login` (Azure CLI authentication)
- V1: AZURE_AI_AGENT_PROJECT_CONNECTION_STRING
- V2: AZURE_AI_PROJECT_ENDPOINT, AZURE_AI_MODEL_DEPLOYMENT_NAME
"""
import asyncio
from collections.abc import AsyncIterable
from contextlib import asynccontextmanager
from collections.abc import AsyncIterator
from agent_framework import (
AgentRunUpdateEvent,
ChatAgent,
HandoffBuilder,
HandoffUserInputRequest,
HostedCodeInterpreterTool,
HostedFileContent,
RequestInfoEvent,
TextContent,
WorkflowEvent,
WorkflowRunState,
WorkflowStatusEvent,
)
from azure.identity.aio import AzureCliCredential
# Toggle between V1 (AzureAIAgentClient) and V2 (AzureAIClient)
USE_V2_CLIENT = False
async def _drain(stream: AsyncIterable[WorkflowEvent]) -> list[WorkflowEvent]:
"""Collect all events from an async stream."""
return [event async for event in stream]
def _handle_events(events: list[WorkflowEvent]) -> tuple[list[RequestInfoEvent], list[str]]:
"""Process workflow events and extract file IDs and pending requests.
Returns:
Tuple of (pending_requests, file_ids_found)
"""
requests: list[RequestInfoEvent] = []
file_ids: list[str] = []
for event in events:
if isinstance(event, WorkflowStatusEvent):
if event.state in {WorkflowRunState.IDLE, WorkflowRunState.IDLE_WITH_PENDING_REQUESTS}:
print(f"[status] {event.state.name}")
elif isinstance(event, RequestInfoEvent):
if isinstance(event.data, HandoffUserInputRequest):
print("\n=== Conversation So Far ===")
for msg in event.data.conversation:
speaker = msg.author_name or msg.role.value
text = msg.text or ""
txt = text[:200] + "..." if len(text) > 200 else text
print(f"- {speaker}: {txt}")
print("===========================\n")
requests.append(event)
elif isinstance(event, AgentRunUpdateEvent):
update = event.data
if update is None:
continue
for content in update.contents:
if isinstance(content, HostedFileContent):
file_ids.append(content.file_id)
print(f"[Found HostedFileContent: file_id={content.file_id}]")
elif isinstance(content, TextContent) and content.annotations:
for annotation in content.annotations:
if hasattr(annotation, "file_id") and annotation.file_id:
file_ids.append(annotation.file_id)
print(f"[Found file annotation: file_id={annotation.file_id}]")
return requests, file_ids
@asynccontextmanager
async def create_agents_v1(credential: AzureCliCredential) -> AsyncIterator[tuple[ChatAgent, ChatAgent]]:
"""Create agents using V1 AzureAIAgentClient."""
from agent_framework.azure import AzureAIAgentClient
async with AzureAIAgentClient(credential=credential) as client:
triage = client.create_agent(
name="triage_agent",
instructions=(
"You are a triage agent. Route code-related requests to the code_specialist. "
"When the user asks to create or generate files, hand off to code_specialist "
"by calling handoff_to_code_specialist."
),
)
code_specialist = client.create_agent(
name="code_specialist",
instructions=(
"You are a Python code specialist. Use the code interpreter to execute Python code "
"and create files when requested. Always save files to /mnt/data/ directory."
),
tools=[HostedCodeInterpreterTool()],
)
yield triage, code_specialist
@asynccontextmanager
async def create_agents_v2(credential: AzureCliCredential) -> AsyncIterator[tuple[ChatAgent, ChatAgent]]:
"""Create agents using V2 AzureAIClient.
Each agent needs its own client instance because the V2 client binds
to a single server-side agent name.
"""
from agent_framework.azure import AzureAIClient
async with (
AzureAIClient(credential=credential) as triage_client,
AzureAIClient(credential=credential) as code_client,
):
triage = triage_client.create_agent(
name="TriageAgent",
instructions=(
"You are a triage agent. Your ONLY job is to route requests to the appropriate specialist. "
"For code or file creation requests, call handoff_to_CodeSpecialist immediately. "
"Do NOT try to complete tasks yourself. Just hand off."
),
)
code_specialist = code_client.create_agent(
name="CodeSpecialist",
instructions=(
"You are a Python code specialist. You have access to a code interpreter tool. "
"Use the code interpreter to execute Python code and create files. "
"Always save files to /mnt/data/ directory. "
"Do NOT discuss handoffs or routing - just complete the coding task directly."
),
tools=[HostedCodeInterpreterTool()],
)
yield triage, code_specialist
async def main() -> None:
"""Run a simple handoff workflow with code interpreter file generation."""
client_version = "V2 (AzureAIClient)" if USE_V2_CLIENT else "V1 (AzureAIAgentClient)"
print(f"=== Handoff Workflow with Code Interpreter File Generation [{client_version}] ===\n")
async with AzureCliCredential() as credential:
create_agents = create_agents_v2 if USE_V2_CLIENT else create_agents_v1
async with create_agents(credential) as (triage, code_specialist):
workflow = (
HandoffBuilder()
.participants([triage, code_specialist])
.set_coordinator(triage)
.with_termination_condition(lambda conv: sum(1 for msg in conv if msg.role.value == "user") >= 2)
.build()
)
user_inputs = [
"Please create a text file called hello.txt with 'Hello from handoff workflow!' inside it.",
"exit",
]
input_index = 0
all_file_ids: list[str] = []
print(f"User: {user_inputs[0]}")
events = await _drain(workflow.run_stream(user_inputs[0]))
requests, file_ids = _handle_events(events)
all_file_ids.extend(file_ids)
input_index += 1
while requests:
request = requests[0]
if input_index >= len(user_inputs):
break
user_input = user_inputs[input_index]
print(f"\nUser: {user_input}")
responses = {request.request_id: user_input}
events = await _drain(workflow.send_responses_streaming(responses))
requests, file_ids = _handle_events(events)
all_file_ids.extend(file_ids)
input_index += 1
print("\n" + "=" * 50)
if all_file_ids:
print(f"SUCCESS: Found {len(all_file_ids)} file ID(s) in handoff workflow:")
for fid in all_file_ids:
print(f" - {fid}")
else:
print("WARNING: No file IDs captured from the handoff workflow.")
print("=" * 50)
"""
Sample Output:
User: Please create a text file called hello.txt with 'Hello from handoff workflow!' inside it.
[Found HostedFileContent: file_id=assistant-JT1sA...]
=== Conversation So Far ===
- user: Please create a text file called hello.txt with 'Hello from handoff workflow!' inside it.
- triage_agent: I am handing off your request to create the text file "hello.txt" with the specified content to the code specialist. They will assist you shortly.
- code_specialist: The file "hello.txt" has been created with the content "Hello from handoff workflow!". You can download it using the link below:
[hello.txt](sandbox:/mnt/data/hello.txt)
===========================
[status] IDLE_WITH_PENDING_REQUESTS
User: exit
[status] IDLE
==================================================
SUCCESS: Found 1 file ID(s) in handoff workflow:
- assistant-JT1sA...
==================================================
""" # noqa: E501
if __name__ == "__main__":
asyncio.run(main())