* Added shell tool
* Fixed CI error
* Add ShellTool support for OpenAI and Anthropic providers
- Add shell_tool_call, shell_tool_result, and shell_command_output content types
- Add ShellTool class and shell_tool decorator to core
- Add get_hosted_shell_tool() to OpenAI Responses client
- Handle shell_call and shell_call_output parsing in OpenAI (sync and streaming)
- Map ShellTool to Anthropic bash tool API format
- Parse bash_code_execution_tool_result as shell_tool_result in Anthropic
- Add unit tests for all new functionality
- Add sample scripts for hosted and local shell execution
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Addressed comments
* Reverted ruff change
* Fixed tests
* Addressed comments
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix IndexError when reasoning models return no text content (#4384)
In _prepare_message_for_openai(), the text_reasoning case unconditionally
accessed all_messages[-1] to attach reasoning_details. When a reasoning
model (e.g. gpt-5-mini) returns reasoning_details without text content,
all_messages is empty, causing an IndexError.
Guard the access by initializing all_messages with the current args dict
when it is empty, so reasoning_details can be safely attached.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: buffer reasoning details for valid message payloads (#4384)
- Buffer pending reasoning details and attach to the next message with
content/tool_calls, avoiding standalone reasoning-only messages.
- When reasoning is the only content, emit a message with empty content
to satisfy Chat Completions schema requirements.
- Strengthen test assertions to verify text+reasoning co-location and
that all messages with reasoning_details also have content or tool_calls.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix text_reasoning handling: always buffer and tighten tests (#4384)
- Always buffer reasoning into pending_reasoning instead of conditionally
attaching to the previous message via fragile all_messages emptiness check
- Attach buffered reasoning to last message at end-of-loop when no subsequent
content consumed it
- Assert exact content values (content == '' not in ('', None))
- Assert exact list lengths (== 1 not >= 1) for stronger regression guards
- Add test for reasoning before FunctionCallContent
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Apply pre-commit auto-fixes
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: handle thread.message.completed event in Assistants API streaming
Previously, `thread.message.completed` events fell through to the
catch-all `else` branch and yielded empty `ChatResponseUpdate` objects,
silently discarding fully-resolved annotation data (file citations,
file paths, and their character-offset regions).
This commit adds a dedicated handler for `thread.message.completed`
that:
- Walks the completed ThreadMessage.content array
- Extracts text blocks with their fully-resolved annotations
- Maps FileCitationAnnotation and FilePathAnnotation to the
framework's Annotation type with proper TextSpanRegion data
- Yields a ChatResponseUpdate containing the complete text and
annotations
Fixes#4322
* test: add tests for thread.message.completed annotation handling
Tests cover:
- File citation annotation extraction
- File path annotation extraction
- Multiple annotations on a single text block
- Text-only messages (no annotations)
- Non-text blocks are skipped
- Mixed content blocks (text + image)
- Conversation ID propagation
* fix: address Copilot review - add quote field and log unrecognized annotations
- Include `quote` from `annotation.file_citation.quote` in
`additional_properties` for FileCitationAnnotation, preserving the
exact cited text snippet from the source file
- Add `else` clause to log unrecognized annotation types at debug level,
consistent with the pattern in `_responses_client.py`
- Add `import logging` and module-level logger
* test: add coverage for quote field and unrecognized annotation logging
- test_message_completed_with_file_citation_quote: verifies quote is
included in additional_properties
- test_message_completed_with_file_citation_no_quote: verifies quote
is omitted when None
- test_message_completed_unrecognized_annotation_logged: verifies
unknown annotation types are logged at debug level and skipped
* fix: address reviewer nits — logger name convention + annotation type string
Per @giles17's review:
- Use logging.getLogger('agent_framework.openai') to match module convention
- Simplify debug message to use annotation.type instead of type().__name__
* refactor: move message.completed tests into consolidated test file
Per @giles17's review: moved all tests from test_assistants_message_completed.py
into test_openai_assistants_client.py and deleted the standalone file.
* fix: resolve mypy no-redef and ruff RET504 lint errors
- Remove duplicate type annotation for 'ann' variable (no-redef)
- Return directly from fixture instead of unnecessary assignment (RET504)
* fix: rename annotation variable in completed block to fix mypy type conflict
The 'annotation' loop variable in thread.message.completed has type
FileCitationAnnotation | FilePathAnnotation, which conflicts with the
delta block's 'annotation' of type FileCitationDeltaAnnotation |
FilePathDeltaAnnotation. Renamed to 'completed_annotation' to avoid
mypy 'Incompatible types in assignment' error.
* fix: remove quote field from FileCitationAnnotation handling
---------
Co-authored-by: Giles Odigwe <79032838+giles17@users.noreply.github.com>
* Fix walrus operator precedence for model_id in AzureOpenAIResponsesClient (#4299)
Add parentheses around the walrus assignment so model_id receives the
actual string value instead of the boolean result of
`kwargs.pop(...) and not deployment_name`.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: replace walrus with explicit None check, add edge-case tests (#4299)
- Replace walrus operator with explicit assignment and 'is not None'
check to avoid boolean-coercion pitfalls (empty string now correctly
surfaces as ValueError instead of silently falling back)
- Add test: deployment_name takes precedence over model_id kwarg
- Add test: model_id='' raises ValueError
- Add test: model_id=None falls back to env var
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add explicit validation for empty model_id in AzureOpenAIResponsesClient
Reject empty or whitespace-only model_id with ValueError instead of
silently passing an empty deployment name downstream. This ensures the
test_init_model_id_kwarg_empty_string test correctly validates behavior
defined in production code rather than relying on downstream validation.
Addresses PR review feedback for #4299.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Simplify model_id handling using walrus operator
Addresses review comment on PR #4310.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Restore explicit model_id validation to fix test failures (#4299)
The walrus operator refactor silently dropped the empty-string validation,
causing test_init_model_id_kwarg_empty_string to fail. Restore the explicit
None check and ValueError raise for empty model_id.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Revert "Restore explicit model_id validation to fix test failures (#4299)"
This reverts commit 1d2965fff6.
* Revert to walrus operator fix per review feedback
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Python: Fix Executor handler type checking with __future__ annotations (#3898)
Use typing.get_type_hints() in _validate_handler_signature to resolve
string annotations from `from __future__ import annotations`. This
mirrors the fix applied to FunctionExecutor in #2308.
When __future__ annotations are enabled, type annotations are stored as
strings. The handler decorator was passing these strings directly to
validate_workflow_context_annotation, which uses typing.get_origin and
returns None for strings, causing a ValueError.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review feedback for #3898: improve error handling and test coverage
- Wrap typing.get_type_hints() in try/except to provide a descriptive
ValueError mentioning the handler name when annotations cannot be resolved
- Strengthen bare context test to assert output_types and workflow_output_types
- Add test for @handler(input=..., output=...) with future annotations
covering the skip_message_annotation branch
- Add test for union-type context annotations with future annotations
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Narrow exception catch and add test for unresolvable annotations (#3898)
- Narrow except clause from bare Exception to (NameError, AttributeError,
TypeError) to avoid masking unexpected errors.
- Add test_handler_unresolvable_annotation_raises to verify that a handler
with a forward-reference to a non-existent type raises ValueError with
the expected message.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix#3898: fall back to raw annotations when get_type_hints fails
When typing.get_type_hints(func) raises NameError (unresolvable forward
ref), AttributeError, RecursionError, or any other exception, fall back
to the raw parameter annotations instead of raising a ValueError.
This matches the suggestion from @moonbox3 on PR #4317.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix test to match new fallback behavior when get_type_hints fails (#3898)
The code now falls back to raw string annotations instead of raising
'Failed to resolve type annotations'. A ValueError is still raised when
the raw string ctx annotation is not a valid WorkflowContext type, so
update the test to match on ValueError without checking the message.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Apply pyupgrade: remove unnecessary string annotation quote
* Add noqa for intentionally undefined name in annotation test
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix _merge_options dropping dict-defined tools (#4303)
_merge_options used getattr(tool, 'name', None) to de-duplicate tools,
which returns None for dict-style tool definitions. This caused all
override dict tools to be treated as duplicates of each other and of any
base dict tools, silently dropping them.
Add _get_tool_name() helper that extracts the name from both object-style
tools (via .name attribute) and dict-style tools (via tool['function']['name']).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review: fix None dedup bug and add comprehensive tests (#4303)
- Exclude None from existing_names set so nameless/malformed tools are
not silently deduplicated against each other
- Add test for cross-type dedup (dict tool + object tool with same name)
- Add test verifying nameless tools are preserved (not falsely deduped)
- Add unit tests for _get_tool_name edge cases: missing function key,
non-dict function value, missing name, no name attribute, non-dict
inputs, and valid dict/object tools
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix OpenAIResponsesClient mishandling single-tool inputs (#4304)
Use normalize_tools() in _prepare_tools_for_openai to wrap single tools
(FunctionTool or dict) in a list before iteration, consistent with the
chat client implementation.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review feedback for #4304
- Use precise type annotation matching normalize_tools/OpenAIChatClient signature
instead of collapsed Sequence[Any] | Any | None
- Move emptiness guard after normalize_tools() call so single falsy tool
objects are not silently swallowed
- Import ToolTypes for the type annotation
- Expand test_prepare_tools_for_openai_single_function_tool assertions to
verify parameters, strict, and parameter schema fields
- Add test_prepare_tools_for_openai_none to verify None input returns []
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
WorkflowAgent._run_impl() and _run_stream_impl() did not set
session_context._response before calling _run_after_providers().
This caused InMemoryHistoryProvider.after_run() to see context.response
as None, so response messages were never stored in the session.
On subsequent runs, the workflow only received prior user inputs without
assistant responses, breaking multi-turn conversations.
Fix: Set session_context._response to the workflow result before running
after_run providers, matching the behavior of the regular Agent class.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
During Assistants API streaming, TextDeltaBlock.text.annotations was
ignored when creating Content objects. This caused raw placeholder
strings like 【4:0†source】 to pass through to downstream consumers
(including AG-UI) instead of being resolved to citation metadata.
Map FileCitationDeltaAnnotation and FilePathDeltaAnnotation from
delta_block.text.annotations to Annotation objects on the Content,
consistent with the existing patterns in _responses_client.py and
_chat_client.py.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix(python): preserve workflow run kwargs on response continuation (#4293)
When continuing a paused workflow with run(responses=...), the existing
run kwargs stored in state were unconditionally overwritten with an empty
dict. This caused subsequent agent invocations to lose the original run
context (e.g., custom_data, user tokens).
Now kwargs are only overwritten when:
- New kwargs are explicitly provided (override), or
- State was just cleared for a fresh run (initialize to {})
On continuation without new kwargs, existing kwargs are preserved.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review feedback for #4293
- Use consistent get_state(key, {}) default pattern in _agent_executor.py
and _workflow_executor.py instead of get_state(key) or {} to safely
handle missing WORKFLOW_RUN_KWARGS_KEY
- Add test for empty-value kwargs on continuation (custom_data={}) to
verify the is-not-None boundary between overwrite and preserve
- Add test for reset_context=True with no kwargs to exercise the elif
branch that initializes WORKFLOW_RUN_KWARGS_KEY to {}
- Add len assertion to override test for consistency
- Document kwargs-collapsing behavior at the public API call site
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Python: Strip reserved kwargs in AgentExecutor to prevent collision (#4295)
workflow.run(session=...) passed 'session' through to agent.run() via
**run_kwargs while AgentExecutor also passes session=self._session
explicitly, causing TypeError: got multiple values for keyword argument.
_prepare_agent_run_args now strips reserved params (session, stream,
messages) from run_kwargs and logs a warning when they are present.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review feedback for #4295
- Use _RESERVED_RUN_PARAMS constant in stripping loop instead of
hardcoded tuple to maintain single source of truth
- Trim frozenset to only stripped keys (session, stream, messages);
options and additional_function_arguments have separate merge logic
- Fix caplog type annotation to use TYPE_CHECKING pattern
- Assert options return value in reserved-kwarg stripping test
- Add test for multiple reserved kwargs supplied simultaneously
- Add integration test for messages= kwarg via workflow.run()
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Python: Fix AgentResponse.value being None when streaming workflow (#3970)
The streaming path in BaseAgent.run() used the raw 'options' parameter
(passed by the caller) to bind response_format into the outer stream's
finalizer. When response_format was set in default_options rather than
runtime options, it was missing from the finalizer and value was None.
Fix: Use the merged chat_options from the run context (via ctx_holder),
matching the non-streaming path which already uses ctx['chat_options'].
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address review feedback for #3970: safer ctx access, add test coverage
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Phase 2: Embedding clients for Ollama, Bedrock, and Azure AI Inference
Add embedding client implementations to existing provider packages:
- OllamaEmbeddingClient: Text embeddings via Ollama's embed API
- BedrockEmbeddingClient: Text embeddings via Amazon Titan on Bedrock
- AzureAIInferenceEmbeddingClient: Text and image embeddings via Azure AI
Inference, supporting Content | str input with separate model IDs for
text (AZURE_AI_INFERENCE_EMBEDDING_MODEL_ID) and image
(AZURE_AI_INFERENCE_IMAGE_EMBEDDING_MODEL_ID) endpoints
Additional changes:
- Rename EmbeddingCoT -> EmbeddingT, EmbeddingOptionsCoT -> EmbeddingOptionsT
- Add otel_provider_name passthrough to all embedding clients
- Register integration pytest marker in all packages
- Add lazy-loading namespace exports for Ollama and Bedrock embeddings
- Add image embedding sample using Cohere-embed-v3-english
- Add azure-ai-inference dependency to azure-ai package
Part of #1188
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix mypy duplicate name and ruff lint issues
- Rename second 'vector' variable to 'img_vector' in image embedding loop
- Combine nested with statements in tests
- Remove unused result assignments in tests
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* updates from feedback
* Fix CI failures in embedding usage handling
- Fix Azure AI embedding mypy issues by normalizing vectors to list[float],
safely accumulating optional usage token fields, and filtering None entries
before constructing GeneratedEmbeddings
- Avoid Bandit false positive by initializing usage details as an empty dict
- Update OpenAI embedding tests to assert canonical usage keys
(input_token_count/total_token_count)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix thread corruption when max_iterations exhausted (#1366)
When the function invocation loop exhausts max_iterations while the model
keeps requesting tools, the failsafe code path (calling the model with
tool_choice='none' and prepending fcc_messages) was unreachable because
'if response is not None: return response' short-circuited before it.
The fix removes the premature return so the failsafe always runs after
loop exhaustion, making a final model call with tool_choice='none' to
produce a clean text answer and prepending accumulated fcc_messages from
prior iterations. This matches the existing pattern used by the error
threshold and max_function_calls paths.
Also unskips test_max_iterations_limit and test_streaming_max_iterations_limit
which were previously skipped with 'needs investigation in unified API'.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add fix report for issue #1366
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix ruff formatting in _tools.py and test_issue_1366_thread_corruption.py
Apply ruff format to fix multi-line string concatenation and function call
formatting issues flagged by the linter.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add quality review for issue #1366 fix
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Remove temporary investigation docs.
* Address PR review: explicit enabled check in log condition, clarify mock behavior in test
- Add explicit function_invocation_configuration['enabled'] check to the
'Maximum iterations reached' log condition in both non-streaming and
streaming paths, making intent clearer when function invocation is disabled.
- Add comment in test_thread_safe_after_max_iterations_with_agent explaining
that the failsafe response (tool_choice='none') is provided automatically
by the mock client, not from run_responses.
* Blend fix and tests into project without issue-specific callouts
- Remove issue #1366 references from _tools.py comments
- Move regression tests from standalone test_issue_1366_thread_corruption.py
into test_function_invocation_logic.py alongside existing max_iterations tests
- Clean up test docstrings to describe behavior generically
- Delete the standalone issue-specific test file
---------
Co-authored-by: alliscode <bentho@microsoft.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* updated integration tests and guidance
* fixed merge test
* updated integration tests
* fix: remove duplicate --dist loadfile flag from pytest-xdist config
Only one --dist mode can be active at a time; the second value silently
overrides the first. Keep --dist worksteal (dynamic load balancing) and
remove the redundant --dist loadfile from all workflow files and
pyproject.toml configs.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* docs: add keep-in-sync notes for merge and integration test workflows
Both python-merge-tests.yml and python-integration-tests.yml share the
same parallel job structure. Added sync reminders in workflow file
comments, the python-testing SKILL.md, and CODING_STANDARD.md.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* refactor: remove RUN_INTEGRATION_TESTS flag
Integration test gating now uses two mechanisms:
- `@pytest.mark.integration` for test selection via `-m` filtering
- `skip_if_*_disabled` for credential/service availability checks
The RUN_INTEGRATION_TESTS env var was redundant since the marker handles
selection and the skip decorators already check for actual credentials.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: sync missing env vars from merge-tests to integration-tests
Add OPENAI_EMBEDDINGS_MODEL_ID and AZURE_OPENAI_EMBEDDING_DEPLOYMENT_NAME
to python-integration-tests.yml to match python-merge-tests.yml.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: remove remaining RUN_INTEGRATION_TESTS from embedding tests and docs
Missed test_openai_embedding_client.py and vector-stores README in the
earlier cleanup.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* set functions tests to 3.10
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* feat(python): Add embedding abstractions and OpenAI implementation (Phase 1)
This PR contains two parts:
1. **Overall migration plan** for porting vector stores and embeddings from
Semantic Kernel to Agent Framework (docs/features/vector-stores-and-embeddings/README.md)
covering all 10 phases from core abstractions through connectors and TextSearch.
2. **Phase 1 implementation** — core embedding abstractions and OpenAI/Azure OpenAI
embedding clients:
Core types (_types.py):
- EmbeddingGenerationOptions TypedDict (total=False)
- Embedding[EmbeddingT] generic class with model_id, dimensions, created_at
- GeneratedEmbeddings[EmbeddingT, EmbeddingOptionsT] list container with options, usage
- EmbeddingInputT (default str) and EmbeddingT (default list[float]) TypeVars
Protocol + base class (_clients.py):
- SupportsGetEmbeddings protocol — Generic[EmbeddingInputT, EmbeddingT, OptionsContraT]
- BaseEmbeddingClient ABC — Generic[EmbeddingInputT, EmbeddingT, OptionsCoT]
Telemetry (observability.py):
- EmbeddingTelemetryLayer with gen_ai.operation.name = "embeddings"
OpenAI implementation (openai/_embedding_client.py):
- RawOpenAIEmbeddingClient, OpenAIEmbeddingClient, OpenAIEmbeddingOptions
- Uses _ensure_client() factory pattern
Azure OpenAI implementation (azure/_embedding_client.py):
- AzureOpenAIEmbeddingClient following AzureOpenAIChatClient pattern
- Supports API key, Entra ID credentials, env var configuration
Tests:
- 47 unit tests for types, protocol, base class, OpenAI, and Azure clients
- 6 integration tests (gated behind RUN_INTEGRATION_TESTS + credentials)
Samples:
- samples/02-agents/embeddings/openai_embeddings.py
- samples/02-agents/embeddings/azure_openai_embeddings.py
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Add AzureOpenAIEmbeddingClient to azure __init__.pyi stub
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* ci: Add embedding env vars to Python integration tests
Map OPENAI_EMBEDDING_MODEL_ID and AZURE_OPENAI_EMBEDDING_DEPLOYMENT_NAME
from GitHub vars to the integration test environment.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Handle base64 encoding_format in OpenAI embedding client
When encoding_format='base64' is used, the OpenAI API returns base64-encoded
floats instead of a JSON array. Decode these automatically to list[float]
so the return type stays consistent regardless of encoding format.
Also adds a unit test for base64 decoding and fixes minor docstring/import issues.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Only record INPUT_TOKENS for embedding telemetry
Embeddings have no output/completion tokens. Remove OUTPUT_TOKENS recording
which was double-counting prompt_tokens via the total_tokens fallback.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Resolve mypy variance error and lint warning
Use contravariant/covariant TypeVars for SupportsGetEmbeddings Protocol.
Combine nested if into single statement in telemetry layer.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Make EmbeddingCoT invariant for mypy compatibility
GeneratedEmbeddings is invariant in its type param, so the Protocol
TypeVar cannot be covariant.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Address PR review - empty values guard, service_url for telemetry
- Add early return for empty values in get_embeddings to avoid unnecessary API calls
- Add service_url() method to RawOpenAIEmbeddingClient for proper telemetry endpoint reporting
- Add test for empty values behavior
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Python: Fix OpenAI chat client compatibility with third-party endpoints and OTel 0.4.14 (#4161)
* Fix system message content sent as list instead of string
Some OpenAI-compatible endpoints (e.g. NVIDIA NIM) reject system messages
when content is a list of content parts. This change flattens system and
developer message content to a plain string in the Chat Completions client.
Fixes https://github.com/microsoft/agent-framework/issues/1407
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix compatibility with opentelemetry-semantic-conventions-ai 0.4.14
Version 0.4.14 removed several LLM_* attributes from SpanAttributes
(LLM_SYSTEM, LLM_REQUEST_MODEL, LLM_RESPONSE_MODEL, LLM_REQUEST_MAX_TOKENS,
LLM_REQUEST_TEMPERATURE, LLM_REQUEST_TOP_P, LLM_TOKEN_TYPE).
Move these to the OtelAttr enum with their well-known gen_ai.* string values
and update all references in observability.py and tests.
Fixes https://github.com/microsoft/agent-framework/issues/4160
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Flatten text-only message content to string for all roles
Extend the system/developer fix to all message roles. Text-only content
lists are now post-processed into plain strings, while multimodal content
(text + images/audio) remains as a list. This fixes compatibility with
OpenAI-like endpoints that cannot deserialize list content (e.g. Foundry
Local's Neutron backend).
Partially fixes https://github.com/microsoft/agent-framework/issues/4084
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix streaming text lost when usage data in same chunk
Some providers (e.g. Gemini) include both usage data and text content
in the same streaming chunk. The early return on chunk.usage caused
text and tool call parsing to be skipped entirely. Remove the early
return and process usage alongside text/tool calls.
Fixes https://github.com/microsoft/agent-framework/issues/3434
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix mypy errors in _chat_client.py
Rename shadowed variable 'args' in system/developer branch to 'sys_args'
and rename loop variable 'content' to 'msg_content' to avoid type conflict.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* reorder imports
* fix: Use OtelAttr.REQUEST_MODEL instead of removed SpanAttributes.LLM_REQUEST_MODEL
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* docs: Add score_threshold to vector store plan
Reference SK .NET PR #13501 for score threshold filtering semantics.
Include score_threshold in SearchOptions from Phase 3.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* docs: Add reference to roji's SK .NET MEVD work for SQL connectors
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: Clear env vars in construction tests to avoid CI leakage
Tests for missing API key / model ID now use monkeypatch.delenv to ensure
env vars from the integration test environment don't prevent the expected
ValueError from being raised.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add max_function_calls to FunctionInvocationConfiguration (#2329)
Add a new per-request max_function_calls setting to FunctionInvocationConfiguration
that limits the total number of individual function invocations across all iterations
within a single get_response call. This complements max_iterations (which limits LLM
roundtrips) by providing a hard cap on actual tool executions regardless of parallelism.
- Add max_function_calls field to FunctionInvocationConfiguration (default: None/unlimited)
- Track cumulative function call count in both streaming and non-streaming tool loops
- Force tool_choice='none' when the limit is reached
- Add validation in normalize_function_invocation_configuration
- Improve docstrings for FunctionInvocationConfiguration, FunctionTool, and @tool
to clarify semantics of max_iterations vs max_function_calls vs max_invocations
- Add tests for parallel calls, single calls, unlimited mode, and config validation
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add sample for controlling total tool executions
Showcases all three mechanisms for limiting tool executions:
1. max_iterations — caps LLM roundtrips
2. max_function_calls — caps total individual function invocations per request
3. max_invocations — lifetime cap on a specific tool instance
Plus a combined scenario demonstrating defense in depth.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Suppress ruff E305/fmt in hosting sample to preserve XML doc tags
The XML snippet tags (# <create_agent> / # </create_agent>) are used for
docs extraction and must stay adjacent to the code they wrap. Both ruff
check (E305) and ruff format add blank lines after the function definition,
pushing the closing tag away. Suppress with ruff: noqa: E305 and fmt: off.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add per-agent tool wrapping scenario to control_total_tool_executions sample
Show that wrapping the same callable with @tool multiple times creates
independent FunctionTool instances with separate invocation counters,
enabling per-agent max_invocations budgets for shared functions.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Clarify max_function_calls is a best-effort limit
The limit is checked after each batch of parallel calls completes, so the
current batch always runs to completion even if it overshoots the limit.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Address PR review: fix docstring reference, clarify best-effort in sample
- Fix malformed Sphinx :attr: role in FunctionTool docstring — use plain
backtick reference instead
- Update sample to say 'best-effort cap' instead of 'hard cap' for
max_function_calls, noting it's checked between iterations
- Parametrize pattern is correct (fixture override, matching existing tests)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* clarify max_invocations limits
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix structured_output propagation in ClaudeAgent
Capture structured_output from ResultMessage in _get_stream() and
propagate it to AgentResponse.value via a custom finalizer. Previously
structured_output was silently discarded, making output_format unusable.
Fixes#4095
* Address review feedback: use value parameter instead of private properties
- Extend AgentResponse.from_updates() to accept optional value parameter
- Remove structured_output yield from _get_stream()
- Update _finalize_response() to pass value via public API
- Update streaming test to use get_final_response()
* Fix mypy errors: add value parameter to from_updates overloads
Add value parameter to both @overload signatures of
AgentResponse.from_updates() so mypy recognizes the argument.
---------
Co-authored-by: Amit Mukherjee <amimukherjee@microsoft.com>
Co-authored-by: Dmytro Struk <13853051+dmytrostruk@users.noreply.github.com>
* fix Workflow.as_agent() streaming regression in ag-ui
* Address PR feedback
* workflows wip
* wip
* wip
* Workflow AG-UI demo
* Fixes for handoff workflow demo
* Fixes to workflows support in AG-UI
* Fixes
* Add headers to some demo files
* Fix comment
* Fixes for store
* Make _input_schema lazy-loaded
* fix mypy
* revert session change to handoff only for now
---------
Co-authored-by: Eduard van Valkenburg <eavanvalkenburg@users.noreply.github.com>
* Fix system message content sent as list instead of string
Some OpenAI-compatible endpoints (e.g. NVIDIA NIM) reject system messages
when content is a list of content parts. This change flattens system and
developer message content to a plain string in the Chat Completions client.
Fixes https://github.com/microsoft/agent-framework/issues/1407
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix compatibility with opentelemetry-semantic-conventions-ai 0.4.14
Version 0.4.14 removed several LLM_* attributes from SpanAttributes
(LLM_SYSTEM, LLM_REQUEST_MODEL, LLM_RESPONSE_MODEL, LLM_REQUEST_MAX_TOKENS,
LLM_REQUEST_TEMPERATURE, LLM_REQUEST_TOP_P, LLM_TOKEN_TYPE).
Move these to the OtelAttr enum with their well-known gen_ai.* string values
and update all references in observability.py and tests.
Fixes https://github.com/microsoft/agent-framework/issues/4160
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Flatten text-only message content to string for all roles
Extend the system/developer fix to all message roles. Text-only content
lists are now post-processed into plain strings, while multimodal content
(text + images/audio) remains as a list. This fixes compatibility with
OpenAI-like endpoints that cannot deserialize list content (e.g. Foundry
Local's Neutron backend).
Partially fixes https://github.com/microsoft/agent-framework/issues/4084
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix streaming text lost when usage data in same chunk
Some providers (e.g. Gemini) include both usage data and text content
in the same streaming chunk. The early return on chunk.usage caused
text and tool call parsing to be skipped entirely. Remove the early
return and process usage alongside text/tool calls.
Fixes https://github.com/microsoft/agent-framework/issues/3434
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix mypy errors in _chat_client.py
Rename shadowed variable 'args' in system/developer branch to 'sys_args'
and rename loop variable 'content' to 'msg_content' to avoid type conflict.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: strip function_call and text_reasoning from cross-agent workflow handoff
When a reasoning model (e.g. gpt-5-mini) runs as Agent 1 in a workflow, its
response includes text_reasoning items (with server-scoped IDs like rs_XXXX)
and function_call items. Forwarding these to Agent 2 in a fresh conversation
caused API errors because the reasoning/call IDs are scoped to the original
stored response context.
Changes:
- Strip 'function_call', 'text_reasoning', 'function_approval_request', and
'function_approval_response' from handoff messages in _agent_executor.py
- Keep 'function_result' so the actual tool output content is preserved for
the next agent's context
- Update unit tests to reflect that function_result messages survive handoff
(messages grow from 2→3: user, tool(result), assistant(summary))
- Fix incorrect test assertions in test_function_invocation_stop_clears_*
that assumed the client layer updates session.service_session_id
- Also fixed _extract_function_calls to search all messages with call_id
deduplication, and the error-limit stop path to submit function_call_output
items before halting (via tool_choice=none cleanup call)
Relates to: https://github.com/microsoft/agent-framework/issues/4047
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fix: reasoning model workflow handoff and history serialization
Fixes multiple related issues when using reasoning models (gpt-5-mini,
gpt-5.2) in multi-agent workflows that chain agents via from_response
or replay full conversation history via AgentExecutorRequest.
## Reasoning items always emitted on output_item.added
When a reasoning model produces encrypted or hidden reasoning (no
visible text), the Responses API still fires a reasoning output item
without any reasoning_text.delta events. Previously no text_reasoning
Content was emitted in that case, making it invisible to downstream
logic. Both the non-streaming (_parse_response_from_openai) and
streaming (output_item.added) paths now always emit at least one
text_reasoning Content — with empty text if no content is available —
so co-occurrence detection and serialization guards work reliably.
## Reasoning items only serialized when paired with a function_call
The Responses API only accepts reasoning items in input when they
directly preceded a function_call in the original response. Sending a
reasoning item that preceded a text response (no tool call) causes:
"reasoning was provided without its required following item"
_prepare_message_for_openai now checks has_function_call per message
and skips text_reasoning serialization when there is no accompanying
function_call.
## summary field is an array, not an object
The reasoning item summary field sent to the Responses API must be an
array of objects ([{"type": "summary_text", "text": ...}]), not a
single object. Fixed _prepare_content_for_openai accordingly.
## service_session_id cleared when explicit history is provided
When a workflow coordinator replays a full conversation (including
function calls from a previous agent run) back to an executor via
AgentExecutorRequest or from_response, the executor's session still
held a service_session_id (previous_response_id) from the prior run.
The API then received the same function-call items twice — once from
previous_response_id (server-stored) and once from the explicit input —
causing: "Duplicate item found with id fc_...".
AgentExecutor.run (when should_respond=True) and from_response now
reset self._session.service_session_id = None before running so that
explicit input is the sole source of conversation context.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* small improvements in text reasoning
* refactor: add reset_service_session to AgentExecutorRequest for explicit history replay
Replace the implicit 'always clear service_session_id when should_respond=True'
with an explicit opt-in field on AgentExecutorRequest.
The old approach used should_respond=True as a proxy for 'full history replay',
but that conflates two distinct intents:
- Orchestrations group chat sends should_respond=True with an empty/single-message
list (not a full replay) — unnecessarily clearing service_session_id.
- HITL / feedback coordinators send the full prior conversation and truly need
a fresh service session ID to avoid duplicate-item API errors.
Changes:
- Add AgentExecutorRequest.reset_service_session: bool = False
- AgentExecutor.run only clears service_session_id when this flag is True
- AgentExecutor.from_response unchanged (always clears; always full conversation)
- Set reset_service_session=True in all full-history-replay call sites:
agents_with_HITL.py, azure_chat_agents_tool_calls_with_feedback.py,
autogen-migration round-robin coordinator, tau2 runner
- Update _FullHistoryReplayCoordinator test helper to pass the flag
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* comment update
* fixes from feedback
* fix test
* reverted changes to agent executor
* fix: remove reset_service_session from tau2 runner
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* two other reverts
* fix sample
---------
Co-authored-by: Giles Odigwe <79032838+giles17@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Python: improve .env precedence and observability samples
- Switch load_settings to explicit precedence: overrides -> explicit .env -> environment -> defaults\n- Raise when env_file_path is provided but missing\n- Update settings docs and tests for new behavior\n- Refresh observability samples and README guidance for env loading options\n\nCloses #3864\n\nCo-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* fixed some imports
* Fix load_settings CI regressions
Allow explicit env_file_path values that exist but are not regular files (for example /dev/null) by checking path existence before dotenv parsing, and restore a dict accumulator with typed return cast to satisfy mypy.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Avoid implicit dotenv in observability
Only load dotenv in observability helpers when env_file_path is explicitly provided, and remove test os.devnull workarounds that are no longer necessary.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix streaming branch in weather override middleware sample
The streaming branch of weather_override_middleware only prefixed the
original weather data via a transform hook instead of replacing the
content with the 'perfect weather' override like the non-streaming
branch does. Replace with a new ResponseStream that yields the override
content as ChatResponseUpdate chunks.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fixed exception handling middleware sample
* Fixed runtime context delegation middleware example
* Fixed multimodal input examples
* Small update
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Add workflow support for Azure Functions
* fix compatability with latest framework changes and add integration tests
* refactor code
* remove white space
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* align help text with actual port used
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* replace instance id with a place holder
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* remove unused import
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* remove redundant typing import and fix SIM115
* fix latest breaking changes
* fix mypy issues
* clean up imports
* define source marker strings as constants
* fix json module name
* refactor _extract_message_content_from_dict
* refactor serialization
* add helper method for error response construction and remove _extract_message_content_from_dict since it is not needed
* use strict tpe checking for edges
* change how duplicate agent registrations are handled
* cancel approval_task on HITL timeout
* update docstring
* fix: align azurefunctions package with core API changes after rebase
- State.import_state/export_state are now sync (removed await)
- Add State.commit() before export_state() in activity execution
- Rename executor parameter shared_state -> state
- Rename ctx.set_shared_state/get_shared_state -> set_state/get_state (sync)
- WorkflowBuilder now takes start_executor as constructor kwarg
- Update WorkflowOutputEvent -> WorkflowEvent with type='output'
- Update RequestInfoEvent -> WorkflowEvent[Any]
- Update SharedState -> State in test imports
- Update duplicate agent name tests to match new warning behavior
- Update sample README API references
* fix sample check errors
* fix mypy issues
* fix trailing white spaces
* fix test imports
* feat: add durable workflow samples and adapt to main branch changes
- Add workflow samples 09-12 to 04-hosting/azure_functions/
- Adapt to ChatMessage -> Message rename from main
- Adapt to pickle-based checkpoint encoding from main
- Simplify _serialization.py to delegate to core encode/decode
- Fix Message -> WorkflowMessage disambiguation in _context.py
- Remove non-existent _checkpoint_summary import
* fix: update create_checkpoint signature to match superclass
* fix: correct relative link in HITL sample README
* fix: resolve import breakage after rebase (State, DurableAgentThread, get_logger)
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Dmytro Struk <13853051+dmytrostruk@users.noreply.github.com>
* feat: Inject OpenTelemetry trace context into MCP requests and update documentation
* Update python/samples/getting_started/observability/README.md
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Update python/packages/core/tests/core/test_mcp.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* refactor: move opentelemetry import to module level
OpenTelemetry is a hard dependency of agent-framework-core (per
pyproject.toml), so the try/except ImportError guard was dead code.
Move the import to the top of the file to fail fast on missing
dependencies instead of silently hiding installation issues.
---------
Co-authored-by: Pete Roden <Pete.Roden@microsoft.com>
Co-authored-by: Mark Wallace <127216156+markwallace-microsoft@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Fix#3600: Pass JSON schemas through without Pydantic conversion
This change optimizes FunctionTool and MCP flows by passing JSON schemas
directly to providers without converting them to Pydantic models first.
Key changes:
- Store JSON schema as-is when supplied to FunctionTool
- Skip Pydantic model_validate for schema-supplied tools in invoke()
- Return MCP tool schemas directly without conversion
- Add comprehensive tests for schema passthrough behavior
Performance benefits:
- Eliminates expensive Pydantic model creation for supplied schemas
- Preserves exact schema structure (additionalProperties, custom fields, etc.)
- Reduces memory overhead and initialization time
Maintains backward compatibility:
- Function signature inference still uses Pydantic models
- Explicit Pydantic models passed as input_model work as before
- All existing tests pass
* Fix schema passthrough validation and remove helper
* Simplify FunctionTool without generic model dependency
* Fix FunctionTool typing fallout in 3600
* Remove FunctionTool[Any] compatibility shim
* Use serializable kwargs in OTEL tool args
* Python: Replace wildcard imports with explicit imports
- Replace all 'from ... import *' with explicit symbol imports
- Add __all__ declarations to namespace packages for re-exports
- Update CODING_STANDARD.md to prohibit wildcard imports
- Maintain exported API and preserve all functionality
fixes#3605
* Refine wildcard guidance example text
* Simplify explicit exports without self-aliases
* fix: prevent repeating instructions in continued Responses API conversations
- Instructions are now only prepended to messages on the first turn
- When conversation_id/response_id exists (continuation), instructions are skipped
- Covers OpenAI and Azure Responses API paths
- Adds regression tests for all continuation scenarios
Fixes#3498
* Apply lint fixes to continuation tests
* Consolidate responses continuation tests
Tao ChenandGitHub
