diff --git a/util/opentelemetry-util-genai/.changelog/17.added b/util/opentelemetry-util-genai/.changelog/17.added
new file mode 100644
index 00000000..5b2f2241
--- /dev/null
+++ b/util/opentelemetry-util-genai/.changelog/17.added
@@ -0,0 +1 @@
+Add `tool_result` as a parameter to `ToolInvocation`, add `tool_result` and `arguments` as span attributes to the `execute_tool` span if `ContentCapture` flag is set.
\ No newline at end of file
diff --git a/util/opentelemetry-util-genai/.changelog/17.deprecated b/util/opentelemetry-util-genai/.changelog/17.deprecated
new file mode 100644
index 00000000..3c55a8eb
--- /dev/null
+++ b/util/opentelemetry-util-genai/.changelog/17.deprecated
@@ -0,0 +1 @@
+deprecate all `start_` factories, update all `invocation` factories to return objects that can be used as ContextManager's
\ No newline at end of file
diff --git a/util/opentelemetry-util-genai/AGENTS.md b/util/opentelemetry-util-genai/AGENTS.md
index 77cc5cf2..4eaa4fc8 100644
--- a/util/opentelemetry-util-genai/AGENTS.md
+++ b/util/opentelemetry-util-genai/AGENTS.md
@@ -25,7 +25,7 @@ applicable ones.
 Every new operation type must follow this pattern:
 
 ```python
-invocation = handler.start_inference(provider, request_model, server_address=..., server_port=...)
+invocation = handler.inference(provider, request_model, server_address=..., server_port=...)
 invocation.temperature = ...
 try:
     response = client.call(...)
@@ -39,24 +39,23 @@ except Exception as exc:
 
 Factory methods on `TelemetryHandler` (`handler.py`):
 
-- `start_inference(provider, request_model, *, server_address, server_port)` → `InferenceInvocation`
-- `start_embedding(provider, request_model, *, server_address, server_port)` → `EmbeddingInvocation`
-- `start_tool(name, *, arguments, tool_call_id, tool_type, tool_description)` → `ToolInvocation`
-- `start_workflow(name)` → `WorkflowInvocation`
+- `inference(provider, request_model, *, server_address, server_port)` → `InferenceInvocation`
+- `embedding(provider, request_model, *, server_address, server_port)` → `EmbeddingInvocation`
+- `tool(name, *, arguments, tool_call_id, tool_type, tool_description)` → `ToolInvocation`
+- `workflow(name)` → `WorkflowInvocation`
 
-Context manager equivalents (`handler.inference()`, `handler.embedding()`, `handler.tool()`,
-`handler.workflow()`) are available when the span lifetime maps cleanly to a `with` block.
+The returned object can also be used as a context manager (`with ... as invocation:`) when the span lifetime maps cleanly to a `with` block.
 
-`start_*()` factories must map 1:1 to distinct semconv operation types (inference, embeddings,
+The above factories must map 1:1 to distinct semconv operation types (inference, embeddings,
 tool execution, agent invocation, workflow invocation). Names must match the operation
 unambiguously — for example, `create_agent` and `invoke_agent` are different operations, so a
-single `start_agent()` would be ambiguous and is not acceptable. Add a new factory per operation
+single `agent()` would be ambiguous and is not acceptable. Add a new factory per operation
 instead.
 
-Factory names are Python-style singular verbs (`start_embedding`, `start_tool`); the op names
-they map to follow semconv (`embeddings`, `tool execution`, future operations).
+Factory names are Python-style singular verbs (`inference`, `embedding`, `tool`, `workflow`); the op names
+they map to follow semconv operations.
 
-`start_*()` factories must accept all attributes that semconv marks as important for sampling
+Factory methods must accept all attributes that semconv marks as important for sampling
 decisions as parameters, so they are on the span at creation time. Attributes that are also
 marked required by semconv must be required parameters (no default value). Operation name
 is usually hardcoded in specific invocation and does not need to be passed.
@@ -65,7 +64,7 @@ is usually hardcoded in specific invocation and does not need to be passed.
 
 **Never construct invocation types directly** (`InferenceInvocation(...)`, `ToolInvocation(...)`,
 etc.) in instrumentation or production code — direct construction skips span creation and context
-propagation, so all telemetry calls become no-ops. Always use `handler.start_*()`.
+propagation, so all telemetry calls become no-ops. Always use `handler.*()`.
 
 ## 3. Exception Handling
 
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_agent_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_agent_invocation.py
index 675c06fa..2afc2780 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_agent_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_agent_invocation.py
@@ -36,9 +36,8 @@
 class AgentInvocation(GenAIInvocation):
     """Represents a single agent invocation (invoke_agent span).
 
-    Use handler.start_invoke_local_agent() / handler.start_invoke_remote_agent()
-    or the handler.invoke_local_agent() / handler.invoke_remote_agent() context
-    managers rather than constructing this directly.
+    Use handler.invoke_local_agent() or handler.invoke_remote_agent()
+    rather than constructing this directly.
 
     Reference:
         Client span: https://github.com/open-telemetry/semantic-conventions/blob/main/docs/gen-ai/gen-ai-agent-spans.md#invoke-agent-client-span
@@ -59,7 +58,7 @@ def __init__(
         server_port: int | None = None,
         agent_name: str | None = None,
     ) -> None:
-        """Use handler.start_invoke_local_agent() or handler.start_invoke_remote_agent() instead of calling this directly."""
+        """Use handler.invoke_local_agent() or handler.invoke_remote_agent() instead of calling this directly."""
         _operation_name = GenAI.GenAiOperationNameValues.INVOKE_AGENT.value
         super().__init__(
             tracer,
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_embedding_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_embedding_invocation.py
index a3f6909e..eedd2fcd 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_embedding_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_embedding_invocation.py
@@ -20,8 +20,7 @@
 class EmbeddingInvocation(GenAIInvocation):
     """Represents a single embedding model invocation.
 
-    Use handler.start_embedding(provider) or the handler.embedding(provider)
-    context manager rather than constructing this directly.
+    Use handler.embedding(provider) rather than constructing this directly.
     """
 
     def __init__(
@@ -36,7 +35,7 @@ def __init__(
         server_address: str | None = None,
         server_port: int | None = None,
     ) -> None:
-        """Use handler.start_embedding(provider) or handler.embedding(provider) instead of calling this directly."""
+        """Use handler.embedding(provider) rather than calling this directly."""
         _operation_name = GenAI.GenAiOperationNameValues.EMBEDDINGS.value
         super().__init__(
             tracer,
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_inference_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_inference_invocation.py
index 017c3e3f..bcfb6692 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_inference_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_inference_invocation.py
@@ -37,8 +37,7 @@
 class InferenceInvocation(GenAIInvocation):
     """Represents a single LLM chat/completion call.
 
-    Use handler.start_inference(provider) or the handler.inference(provider)
-    context manager rather than constructing this directly.
+    Use handler.inference(provider) rather than constructing this directly.
     """
 
     def __init__(
@@ -57,7 +56,7 @@ def __init__(
         operation_name = (
             operation_name or GenAI.GenAiOperationNameValues.CHAT.value
         )
-        """Use handler.start_inference(provider) or handler.inference(provider) instead of calling this directly."""
+        """Use handler.inference(provider) rather than calling this directly."""
         super().__init__(
             tracer,
             metrics_recorder,
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_invocation.py
index 14efd7ac..36cc3849 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_invocation.py
@@ -4,13 +4,14 @@
 from __future__ import annotations
 
 import timeit
-from abc import ABC, abstractmethod
-from contextlib import contextmanager
+from abc import abstractmethod
+from contextlib import AbstractContextManager
 from contextvars import Token
 from dataclasses import asdict
-from typing import TYPE_CHECKING, Any, Generator, Sequence
+from types import TracebackType
+from typing import TYPE_CHECKING, Any, Sequence
 
-from typing_extensions import Self, TypeAlias
+from typing_extensions import TypeAlias
 
 from opentelemetry._logs import Logger, LogRecord
 from opentelemetry.context import Context, attach, detach
@@ -39,10 +40,11 @@
 if TYPE_CHECKING:
     from opentelemetry.util.genai.metrics import InvocationMetricsRecorder
 
+
 ContextToken: TypeAlias = Token[Context]
 
 
-class GenAIInvocation(ABC):
+class GenAIInvocation(AbstractContextManager["GenAIInvocation"]):
     """
     Base class for all GenAI invocation types. Manages the lifecycle of a single
     GenAI operation (LLM call, embedding, tool execution, workflow, etc.).
@@ -165,15 +167,19 @@ def fail(self, error: Error | BaseException) -> None:
             error = Error(type=type(error), message=str(error))
         self._finish(error)
 
-    @contextmanager
-    def _managed(self) -> Generator[Self, None, None]:
-        """Context manager that calls stop() on success or fail() on exception."""
-        try:
-            yield self
-        except Exception as exc:
-            self.fail(exc)
-            raise
-        self.stop()
+    def __enter__(self) -> GenAIInvocation:
+        return self
+
+    def __exit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        traceback: TracebackType | None,
+    ) -> None:
+        if exc_value is not None and isinstance(exc_value, Exception):
+            self.fail(exc_value)
+        else:
+            self.stop()
 
 
 def get_content_attributes(
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_tool_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_tool_invocation.py
index 2aa65140..d35cad3a 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_tool_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_tool_invocation.py
@@ -13,6 +13,10 @@
 from opentelemetry.util.genai._invocation import Error, GenAIInvocation
 from opentelemetry.util.genai.completion_hook import CompletionHook
 from opentelemetry.util.genai.metrics import InvocationMetricsRecorder
+from opentelemetry.util.genai.utils import (
+    should_capture_content_on_spans,
+)
+from opentelemetry.util.types import AttributeValue
 
 
 class ToolInvocation(GenAIInvocation):
@@ -43,10 +47,11 @@ def __init__(
         completion_hook: CompletionHook,
         name: str,
         *,
-        arguments: Any = None,
+        arguments: AttributeValue | None = None,
         tool_call_id: str | None = None,
         tool_type: str | None = None,
         tool_description: str | None = None,
+        tool_result: AttributeValue | None = None,
     ) -> None:
         """Use handler.start_tool(name) or handler.tool(name) instead of calling this directly."""
         _operation_name = GenAI.GenAiOperationNameValues.EXECUTE_TOOL.value
@@ -58,12 +63,13 @@ def __init__(
             operation_name=_operation_name,
             span_name=f"{_operation_name} {name}" if name else _operation_name,
         )
+        self.should_capture_content_on_span = should_capture_content_on_spans()
         self.name = name
+        self.tool_result = tool_result
         self.arguments = arguments
         self.tool_call_id = tool_call_id
         self.tool_type = tool_type
         self.tool_description = tool_description
-        self.tool_result: Any = None
         self._start(self._get_base_attributes())
 
     def _get_base_attributes(self) -> dict[str, Any]:
@@ -73,6 +79,18 @@ def _get_base_attributes(self) -> dict[str, Any]:
             (GenAI.GEN_AI_TOOL_CALL_ID, self.tool_call_id),
             (GenAI.GEN_AI_TOOL_TYPE, self.tool_type),
             (GenAI.GEN_AI_TOOL_DESCRIPTION, self.tool_description),
+            (
+                GenAI.GEN_AI_TOOL_CALL_ARGUMENTS,
+                self.arguments
+                if self.should_capture_content_on_span
+                else None,
+            ),
+            (
+                GenAI.GEN_AI_TOOL_CALL_RESULT,
+                self.tool_result
+                if self.should_capture_content_on_span
+                else None,
+            ),
         )
         return {
             GenAI.GEN_AI_OPERATION_NAME: self._operation_name,
@@ -94,7 +112,18 @@ def _apply_finish(self, error: Error | None = None) -> None:
             (GenAI.GEN_AI_TOOL_CALL_ID, self.tool_call_id),
             (GenAI.GEN_AI_TOOL_TYPE, self.tool_type),
             (GenAI.GEN_AI_TOOL_DESCRIPTION, self.tool_description),
-            (GenAI.GEN_AI_TOOL_CALL_ARGUMENTS, self.arguments),
+            (
+                GenAI.GEN_AI_TOOL_CALL_ARGUMENTS,
+                self.arguments
+                if self.should_capture_content_on_span
+                else None,
+            ),
+            (
+                GenAI.GEN_AI_TOOL_CALL_RESULT,
+                self.tool_result
+                if self.should_capture_content_on_span
+                else None,
+            ),
         )
         attributes: dict[str, Any] = {
             GenAI.GEN_AI_OPERATION_NAME: self._operation_name,
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_workflow_invocation.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_workflow_invocation.py
index 9d30ab3a..a72dfb58 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_workflow_invocation.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/_workflow_invocation.py
@@ -32,8 +32,7 @@ class WorkflowInvocation(GenAIInvocation):
     and retrieval invocations). A workflow groups multiple operations together,
     accepting input(s) and producing final output(s).
 
-    Use handler.start_workflow(name) or the handler.workflow(name) context
-    manager rather than constructing this directly.
+    Use handler.workflow(name) rather than constructing this directly.
     """
 
     def __init__(
@@ -44,7 +43,7 @@ def __init__(
         completion_hook: CompletionHook,
         name: str | None,
     ) -> None:
-        """Use handler.start_workflow(name) or handler.workflow(name) instead of calling this directly."""
+        """Use handler.workflow(name) rather than calling this directly."""
         _operation_name = "invoke_workflow"
         super().__init__(
             tracer,
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/handler.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/handler.py
index 9839e041..3ab7335c 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/handler.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/handler.py
@@ -17,7 +17,7 @@
     handler = get_telemetry_handler()
 
     # Factory method: construct and start in one call, then stop or fail.
-    invocation = handler.start_inference("my-provider", request_model="my-model")
+    invocation = handler.inference("my-provider", request_model="my-model")
     invocation.input_messages = [...]
     invocation.temperature = 0.7
     try:
@@ -38,7 +38,6 @@
 from __future__ import annotations
 
 import os
-from contextlib import AbstractContextManager
 
 from opentelemetry._logs import (
     LoggerProvider,
@@ -74,6 +73,7 @@
     is_experimental_mode,
 )
 from opentelemetry.util.genai.version import __version__
+from opentelemetry.util.types import AttributeValue
 
 
 class TelemetryHandler:
@@ -145,6 +145,9 @@ def start_inference(
     ) -> InferenceInvocation:
         """Create and start an LLM inference invocation.
 
+        .. deprecated:: 0.5b0
+            Use ``handler.inference()`` instead.
+
         Set remaining attributes (input_messages, temperature, etc.) on the
         returned invocation, then call invocation.stop() or invocation.fail().
         """
@@ -164,7 +167,7 @@ def start_llm(self, invocation: LLMInvocation) -> LLMInvocation:
         """Start an LLM invocation.
 
         .. deprecated::
-            Use ``handler.start_inference()`` instead.
+            Use ``handler.inference()`` instead.
         """
         invocation._start_with_handler(
             self._tracer,
@@ -184,6 +187,9 @@ def start_embedding(
     ) -> EmbeddingInvocation:
         """Create and start an Embedding invocation.
 
+        .. deprecated:: 0.5b0
+            Use ``handler.embedding()`` instead.
+
         Set remaining attributes (encoding_formats, etc.) on the returned
         invocation, then call invocation.stop() or invocation.fail().
         """
@@ -202,13 +208,16 @@ def start_tool(
         self,
         name: str,
         *,
-        arguments: object = None,
+        arguments: AttributeValue | None = None,
         tool_call_id: str | None = None,
         tool_type: str | None = None,
         tool_description: str | None = None,
     ) -> ToolInvocation:
         """Create and start a tool invocation.
 
+        .. deprecated:: 0.5b0
+            Use ``handler.tool()`` instead.
+
         Set tool_result on the returned invocation when done, then call
         invocation.stop() or invocation.fail().
         """
@@ -231,6 +240,9 @@ def start_workflow(
     ) -> WorkflowInvocation:
         """Create and start a workflow invocation.
 
+        .. deprecated:: 0.5b0
+            Use ``handler.workflow()`` instead.
+
         Set remaining attributes on the returned invocation, then call
         invocation.stop() or invocation.fail().
         """
@@ -246,7 +258,7 @@ def stop_llm(self, invocation: LLMInvocation) -> LLMInvocation:  # pylint: disab
         """Finalize an LLM invocation successfully and end its span.
 
         .. deprecated::
-            Use ``handler.start_inference()``  and then ``inference.stop()`` instead.
+            Use ``handler.inference()``  and then ``inference.stop()`` instead.
         """
         invocation._sync_to_invocation()
         if invocation._inference_invocation is not None:
@@ -261,7 +273,7 @@ def fail_llm(  # pylint: disable=no-self-use
         """Fail an LLM invocation and end its span with error status.
 
         .. deprecated::
-            Use ``handler.start_inference()``  and then ``inference.fail()`` instead.
+            Use ``handler.inference()``  and then ``inference.fail()`` instead.
         """
         invocation._sync_to_invocation()
         if invocation._inference_invocation is not None:
@@ -275,21 +287,27 @@ def inference(
         request_model: str | None = None,
         server_address: str | None = None,
         server_port: int | None = None,
-    ) -> AbstractContextManager[InferenceInvocation]:
-        """Context manager for LLM inference invocations.
+        operation_name: str | None = None,
+    ) -> InferenceInvocation:
+        """Returns an Inference invocation. Starts span when called.
 
-        Only set data attributes on the invocation object, do not modify the span or context.
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
+        Only set data attributes on the invocation object, do not modify the span or context.
         """
-        return self.start_inference(
+        return InferenceInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
             provider=provider,
             request_model=request_model,
             server_address=server_address,
             server_port=server_port,
-        )._managed()
+            operation_name=operation_name,
+        )
 
     def embedding(
         self,
@@ -298,46 +316,54 @@ def embedding(
         request_model: str | None = None,
         server_address: str | None = None,
         server_port: int | None = None,
-    ) -> AbstractContextManager[EmbeddingInvocation]:
-        """Context manager for Embedding invocations.
+    ) -> EmbeddingInvocation:
+        """Returns an Embedding invocation. Starts span when called.
 
-        Only set data attributes on the invocation object, do not modify the span or context.
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
+        Only set data attributes on the invocation object, do not modify the span or context.
         """
-        return self.start_embedding(
+        return EmbeddingInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
             provider=provider,
             request_model=request_model,
             server_address=server_address,
             server_port=server_port,
-        )._managed()
+        )
 
     def tool(
         self,
         name: str,
         *,
-        arguments: object = None,
+        arguments: AttributeValue | None = None,
         tool_call_id: str | None = None,
         tool_type: str | None = None,
         tool_description: str | None = None,
-    ) -> AbstractContextManager[ToolInvocation]:
-        """Context manager for Tool invocations.
+    ) -> ToolInvocation:
+        """Returns a Tool invocation. Starts span when called.
 
-        Only set data attributes on the invocation object, do not modify the span or context.
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
+        Only set data attributes on the invocation object, do not modify the span or context.
         """
-        return self.start_tool(
+        return ToolInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
             name,
             arguments=arguments,
             tool_call_id=tool_call_id,
             tool_type=tool_type,
             tool_description=tool_description,
-        )._managed()
+        )
 
     def start_invoke_local_agent(
         self,
@@ -348,6 +374,9 @@ def start_invoke_local_agent(
     ) -> AgentInvocation:
         """Create and start a local agent invocation (INTERNAL span kind).
 
+        .. deprecated:: 0.5b0
+            Use ``handler.invoke_local_agent()`` instead.
+
         Use for agents running within the same process (e.g. LangChain, CrewAI).
 
         Set remaining attributes (agent_name, etc.) on the returned invocation,
@@ -375,6 +404,9 @@ def start_invoke_remote_agent(
     ) -> AgentInvocation:
         """Create and start a remote agent invocation (CLIENT span kind).
 
+        .. deprecated:: 0.5b0
+            Use ``handler.invoke_remote_agent()`` instead.
+
         Use for agents invoked over a remote service (e.g. OpenAI Assistants, AWS Bedrock).
 
         Set remaining attributes (agent_name, etc.) on the returned invocation,
@@ -399,22 +431,27 @@ def invoke_local_agent(
         *,
         request_model: str | None = None,
         agent_name: str | None = None,
-    ) -> AbstractContextManager[AgentInvocation]:
-        """Context manager for local agent invocations (INTERNAL span kind).
+    ) -> AgentInvocation:
+        """Returns an agent invocation (INTERNAL span kind). Starts span when called.
+
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
         Use for agents running within the same process (e.g. LangChain, CrewAI).
 
         Only set data attributes on the invocation object, do not modify the span or context.
-
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
         """
-        return self.start_invoke_local_agent(
+        return AgentInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
             provider,
+            span_kind=SpanKind.INTERNAL,
             request_model=request_model,
             agent_name=agent_name,
-        )._managed()
+        )
 
     def invoke_remote_agent(
         self,
@@ -424,38 +461,49 @@ def invoke_remote_agent(
         server_address: str | None = None,
         server_port: int | None = None,
         agent_name: str | None = None,
-    ) -> AbstractContextManager[AgentInvocation]:
-        """Context manager for remote agent invocations (CLIENT span kind).
+    ) -> AgentInvocation:
+        """Returns an agent invocation (CLIENT span kind). Starts span when called.
+
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
         Use for agents invoked over a remote service (e.g. OpenAI Assistants, AWS Bedrock).
 
         Only set data attributes on the invocation object, do not modify the span or context.
-
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
         """
-        return self.start_invoke_remote_agent(
+        return AgentInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
             provider,
+            span_kind=SpanKind.CLIENT,
             request_model=request_model,
             agent_name=agent_name,
             server_address=server_address,
             server_port=server_port,
-        )._managed()
+        )
 
     def workflow(
         self,
         name: str | None = None,
-    ) -> AbstractContextManager[WorkflowInvocation]:
-        """Context manager for Workflow invocations.
+    ) -> WorkflowInvocation:
+        """Returns a Workflow invocation. Starts a span when called.
 
-        Only set data attributes on the invocation object, do not modify the span or context.
+        Returned object can be used as a ContextManager which automatically calls `stop` or `fail`
+        to finalize the span upon exiting. If not used as a ContextManager, the caller is
+        responsible for calling `stop` or `fail` to finalize the span.
 
-        Starts the span on entry. On normal exit, finalizes the invocation and ends the span.
-        If an exception occurs inside the context, marks the span as error, ends it, and
-        re-raises the original exception.
+        Only set data attributes on the invocation object, do not modify the span or context.
         """
-        return self.start_workflow(name=name)._managed()
+        return WorkflowInvocation(
+            self._tracer,
+            self._metrics_recorder,
+            self._logger,
+            self._completion_hook,
+            name,
+        )
 
 
 def get_telemetry_handler(
diff --git a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/utils.py b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/utils.py
index 67efc227..eec6906a 100644
--- a/util/opentelemetry-util-genai/src/opentelemetry/util/genai/utils.py
+++ b/util/opentelemetry-util-genai/src/opentelemetry/util/genai/utils.py
@@ -112,6 +112,17 @@ def should_capture_content_on_spans_in_experimental_mode() -> bool:
     return True
 
 
+def should_capture_content_on_spans() -> bool:
+    "Returns whether capture content is enabled regardless of which mode."
+    if is_experimental_mode():
+        return get_content_capturing_mode() in (
+            ContentCapturingMode.SPAN_ONLY,
+            ContentCapturingMode.SPAN_AND_EVENT,
+        )
+    envvar = os.environ.get(OTEL_INSTRUMENTATION_GENAI_CAPTURE_MESSAGE_CONTENT)
+    return bool(envvar) and envvar.lower().strip() == "true"
+
+
 class _GenAiJsonEncoder(json.JSONEncoder):
     def default(self, o: Any) -> Any:
         if isinstance(o, bytes):
diff --git a/util/opentelemetry-util-genai/tests/test_handler_agent.py b/util/opentelemetry-util-genai/tests/test_handler_agent.py
index 9f1179e1..aa4812d3 100644
--- a/util/opentelemetry-util-genai/tests/test_handler_agent.py
+++ b/util/opentelemetry-util-genai/tests/test_handler_agent.py
@@ -39,7 +39,7 @@ def setUp(self):
         self.handler = TelemetryHandler(tracer_provider=tracer_provider)
 
     def test_start_stop_creates_span(self):
-        invocation = self.handler.start_invoke_local_agent(
+        invocation = self.handler.invoke_local_agent(
             "openai",
             request_model="gpt-4",
             agent_name="Math Tutor",
@@ -56,7 +56,7 @@ def test_start_stop_creates_span(self):
         assert span.attributes[GenAI.GEN_AI_REQUEST_MODEL] == "gpt-4"
 
     def test_span_kind_internal(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.stop()
         assert (
             self.span_exporter.get_finished_spans()[0].kind
@@ -64,14 +64,14 @@ def test_span_kind_internal(self):
         )
 
     def test_no_server_attributes(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.stop()
         attrs = self.span_exporter.get_finished_spans()[0].attributes
         assert server_attributes.SERVER_ADDRESS not in attrs
         assert server_attributes.SERVER_PORT not in attrs
 
     def test_all_attributes(self):
-        invocation = self.handler.start_invoke_local_agent(
+        invocation = self.handler.invoke_local_agent(
             "openai",
             request_model="gpt-4",
         )
@@ -116,7 +116,7 @@ def test_all_attributes(self):
         assert attrs[GenAI.GEN_AI_RESPONSE_FINISH_REASONS] == ("stop",)
 
     def test_finish_reasons_multiple(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.finish_reasons = ["stop", "length"]
         invocation.stop()
         attrs = self.span_exporter.get_finished_spans()[0].attributes
@@ -126,7 +126,7 @@ def test_finish_reasons_multiple(self):
         )
 
     def test_finish_reasons_empty_list_omitted(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.finish_reasons = []
         invocation.stop()
         attrs = self.span_exporter.get_finished_spans()[0].attributes
@@ -134,7 +134,7 @@ def test_finish_reasons_empty_list_omitted(self):
         assert GenAI.GEN_AI_RESPONSE_FINISH_REASONS not in attrs
 
     def test_cache_token_attributes(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.input_tokens = 100
         invocation.cache_creation_input_tokens = 25
         invocation.cache_read_input_tokens = 50
@@ -146,7 +146,7 @@ def test_cache_token_attributes(self):
         assert attrs[GenAI.GEN_AI_USAGE_CACHE_READ_INPUT_TOKENS] == 50
 
     def test_fail_sets_error_status(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.fail(RuntimeError("agent crashed"))
 
         span = self.span_exporter.get_finished_spans()[0]
@@ -183,7 +183,7 @@ def test_context_manager_default_invocation(self):
         assert len(self.span_exporter.get_finished_spans()) == 1
 
     def test_default_values(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.stop()
         assert invocation._operation_name == "invoke_agent"
         assert invocation.agent_name is None
@@ -198,7 +198,7 @@ def test_default_values(self):
         assert not invocation.attributes
 
     def test_with_messages(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.input_messages = [
             InputMessage(role="user", parts=[Text(content="Hello")])
         ]
@@ -214,7 +214,7 @@ def test_with_messages(self):
         assert invocation.input_messages[0].role == "user"
 
     def test_custom_attributes(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.attributes["custom.key"] = "custom_value"
         invocation.stop()
         spans = self.span_exporter.get_finished_spans()
@@ -226,7 +226,7 @@ def test_tool_definitions_type(self):
             description="Get the weather",
             parameters={"type": "object", "properties": {}},
         )
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.tool_definitions = [tool]
         invocation.stop()
         assert len(invocation.tool_definitions) == 1
@@ -234,23 +234,23 @@ def test_tool_definitions_type(self):
         assert invocation.tool_definitions[0].type == "function"
 
     def test_default_lists_are_independent(self):
-        inv1 = self.handler.start_invoke_local_agent("openai")
-        inv2 = self.handler.start_invoke_local_agent("openai")
+        inv1 = self.handler.invoke_local_agent("openai")
+        inv2 = self.handler.invoke_local_agent("openai")
         inv1.input_messages.append(InputMessage(role="user", parts=[]))
         assert len(inv2.input_messages) == 0
         inv2.stop()
         inv1.stop()
 
     def test_default_attributes_are_independent(self):
-        inv1 = self.handler.start_invoke_local_agent("openai")
-        inv2 = self.handler.start_invoke_local_agent("openai")
+        inv1 = self.handler.invoke_local_agent("openai")
+        inv2 = self.handler.invoke_local_agent("openai")
         inv1.attributes["foo"] = "bar"
         assert "foo" not in inv2.attributes
         inv2.stop()
         inv1.stop()
 
     def test_agent_name_set_after_construction(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.agent_name = "Named Agent"
         invocation.stop()
         span = self.span_exporter.get_finished_spans()[0]
@@ -260,7 +260,7 @@ def test_agent_name_set_after_construction(self):
         assert span.attributes[GenAI.GEN_AI_AGENT_NAME] == "Named Agent"
 
     def test_agent_name_passed_at_construction(self):
-        invocation = self.handler.start_invoke_local_agent(
+        invocation = self.handler.invoke_local_agent(
             "openai", agent_name="Constructor Agent"
         )
         invocation.stop()
@@ -293,7 +293,7 @@ def get_description(self):
         )
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_invoke_local_agent(
+        invocation = handler.invoke_local_agent(
             "openai", agent_name="Sampler Agent"
         )
         invocation.stop()
@@ -325,7 +325,7 @@ def get_description(self):
         )
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_invoke_local_agent("openai")
+        invocation = handler.invoke_local_agent("openai")
         invocation.stop()
 
         assert GenAI.GEN_AI_AGENT_NAME not in captured_attributes
@@ -349,7 +349,7 @@ def setUp(self):
         return_value=True,
     )
     def test_system_instruction_on_span(self, _mock_exp, _mock_cap):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.system_instruction = [
             Text(content="You are a helpful assistant."),
         ]
@@ -372,7 +372,7 @@ def test_tool_definitions_on_span(self, _mock_exp, _mock_cap):
             description="Get the weather",
             parameters={"type": "object", "properties": {}},
         )
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.tool_definitions = [tool]
         invocation.stop()
 
@@ -388,7 +388,7 @@ def test_tool_definitions_on_span(self, _mock_exp, _mock_cap):
         return_value=True,
     )
     def test_messages_on_span(self, _mock_exp, _mock_cap):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.input_messages = [
             InputMessage(role="user", parts=[Text(content="Hello")])
         ]
@@ -406,7 +406,7 @@ def test_messages_on_span(self, _mock_exp, _mock_cap):
         assert GenAI.GEN_AI_OUTPUT_MESSAGES in attrs
 
     def test_content_not_on_span_by_default(self):
-        invocation = self.handler.start_invoke_local_agent("openai")
+        invocation = self.handler.invoke_local_agent("openai")
         invocation.system_instruction = [
             Text(content="You are a helpful assistant."),
         ]
@@ -430,14 +430,14 @@ def setUp(self):
         self.handler = TelemetryHandler(tracer_provider=tracer_provider)
 
     def test_span_kind_client(self):
-        invocation = self.handler.start_invoke_remote_agent("openai")
+        invocation = self.handler.invoke_remote_agent("openai")
         invocation.stop()
         assert (
             self.span_exporter.get_finished_spans()[0].kind == SpanKind.CLIENT
         )
 
     def test_server_attributes(self):
-        invocation = self.handler.start_invoke_remote_agent(
+        invocation = self.handler.invoke_remote_agent(
             "openai",
             server_address="api.openai.com",
             server_port=443,
@@ -448,7 +448,7 @@ def test_server_attributes(self):
         assert attrs[server_attributes.SERVER_PORT] == 443
 
     def test_all_attributes(self):
-        invocation = self.handler.start_invoke_remote_agent(
+        invocation = self.handler.invoke_remote_agent(
             "openai",
             request_model="gpt-4",
             server_address="api.openai.com",
@@ -472,7 +472,7 @@ def test_all_attributes(self):
         assert attrs[GenAI.GEN_AI_REQUEST_MODEL] == "gpt-4"
 
     def test_fail_sets_error_status(self):
-        invocation = self.handler.start_invoke_remote_agent("openai")
+        invocation = self.handler.invoke_remote_agent("openai")
         invocation.fail(RuntimeError("remote agent crashed"))
 
         span = self.span_exporter.get_finished_spans()[0]
@@ -532,7 +532,7 @@ def get_description(self):
         )
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_invoke_remote_agent(
+        invocation = handler.invoke_remote_agent(
             "test-provider",
             request_model="agent-model",
             agent_name="Math Tutor",
@@ -563,7 +563,7 @@ def test_local_agent_records_duration_and_tokens(self) -> None:
             meter_provider=self.meter_provider,
         )
         with patch("timeit.default_timer", return_value=1000.0):
-            invocation = handler.start_invoke_local_agent(
+            invocation = handler.invoke_local_agent(
                 "prov", request_model="model"
             )
         invocation.input_tokens = 5
@@ -612,7 +612,7 @@ def test_remote_agent_records_duration_with_server_attrs(self) -> None:
             tracer_provider=self.tracer_provider,
             meter_provider=self.meter_provider,
         )
-        invocation = handler.start_invoke_remote_agent(
+        invocation = handler.invoke_remote_agent(
             "prov",
             request_model="model",
             server_address="agent.example.com",
@@ -635,7 +635,7 @@ def test_fail_agent_records_error_metric(self) -> None:
             meter_provider=self.meter_provider,
         )
         with patch("timeit.default_timer", return_value=2000.0):
-            invocation = handler.start_invoke_local_agent(
+            invocation = handler.invoke_local_agent(
                 "", request_model="err-model"
             )
         invocation.input_tokens = 11
diff --git a/util/opentelemetry-util-genai/tests/test_handler_completion_hook.py b/util/opentelemetry-util-genai/tests/test_handler_completion_hook.py
index 8ed714b0..19bd874a 100644
--- a/util/opentelemetry-util-genai/tests/test_handler_completion_hook.py
+++ b/util/opentelemetry-util-genai/tests/test_handler_completion_hook.py
@@ -79,7 +79,7 @@ def test_hook_called_on_stop(self):
             )
         ]
 
-        invocation = handler.start_inference("openai", request_model="gpt-4o")
+        invocation = handler.inference("openai", request_model="gpt-4o")
         invocation.input_messages = input_messages
         invocation.output_messages = output_messages
         invocation.system_instruction = system_instruction
@@ -102,7 +102,7 @@ def test_hook_called_on_fail(self):
             InputMessage(role="user", parts=[Text(content="hello")])
         ]
 
-        invocation = handler.start_inference("openai", request_model="gpt-4o")
+        invocation = handler.inference("openai", request_model="gpt-4o")
         invocation.input_messages = input_messages
         invocation.fail(ValueError("boom"))
 
@@ -114,7 +114,7 @@ def test_hook_called_on_fail(self):
     def test_hook_not_called_when_not_set(self):
         # No hook — stop should not raise
         handler = self._make_handler()
-        handler.start_inference("openai", request_model="gpt-4o").stop()
+        handler.inference("openai", request_model="gpt-4o").stop()
 
     def test_log_record_is_none_when_events_disabled(self):
         # Default env: no experimental mode, so log_record should be None.
@@ -123,7 +123,7 @@ def test_log_record_is_none_when_events_disabled(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        handler.start_inference("openai", request_model="gpt-4o").stop()
+        handler.inference("openai", request_model="gpt-4o").stop()
 
         kwargs = hook.on_completion.call_args.kwargs
         self.assertIsNone(kwargs["log_record"])
@@ -137,7 +137,7 @@ def test_log_record_passed_when_events_enabled(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        handler.start_inference("openai", request_model="gpt-4o").stop()
+        handler.inference("openai", request_model="gpt-4o").stop()
 
         kwargs = hook.on_completion.call_args.kwargs
         self.assertIsNotNone(kwargs["log_record"])
@@ -162,7 +162,7 @@ def stamp_ref(*, log_record, **kwargs):
         hook = MagicMock(on_completion=stamp_ref)
         handler = self._make_handler(hook)
 
-        handler.start_inference("openai", request_model="gpt-4o").stop()
+        handler.inference("openai", request_model="gpt-4o").stop()
 
         # The record the hook stamped is the same one that would be emitted
         self.assertIsNotNone(stamped_record)
@@ -281,7 +281,7 @@ def test_workflow_hook_called_on_stop_with_messages(self):
             )
         ]
 
-        invocation = handler.start_workflow(name="my-workflow")
+        invocation = handler.workflow(name="my-workflow")
         invocation.input_messages = input_messages
         invocation.output_messages = output_messages
         invocation.stop()
@@ -300,7 +300,7 @@ def test_workflow_hook_called_on_fail(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        invocation = handler.start_workflow(name="my-workflow")
+        invocation = handler.workflow(name="my-workflow")
         invocation.input_messages = [
             InputMessage(role="user", parts=[Text(content="hello")])
         ]
@@ -314,7 +314,7 @@ def test_workflow_hook_called_with_empty_messages_when_none_set(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        handler.start_workflow(name="my-workflow").stop()
+        handler.workflow(name="my-workflow").stop()
 
         hook.on_completion.assert_called_once()
         kwargs = hook.on_completion.call_args.kwargs
@@ -344,7 +344,7 @@ def test_local_agent_hook_called_on_stop_with_messages(self):
             )
         ]
 
-        invocation = handler.start_invoke_local_agent(
+        invocation = handler.invoke_local_agent(
             "openai", request_model="gpt-4"
         )
         invocation.agent_name = "Math Tutor"
@@ -367,7 +367,7 @@ def test_local_agent_hook_called_on_fail(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        invocation = handler.start_invoke_local_agent(
+        invocation = handler.invoke_local_agent(
             "openai", request_model="gpt-4"
         )
         invocation.input_messages = [
@@ -402,7 +402,7 @@ def test_remote_agent_hook_called_on_stop_with_messages(self):
             )
         ]
 
-        invocation = handler.start_invoke_remote_agent(
+        invocation = handler.invoke_remote_agent(
             "openai",
             request_model="gpt-4",
             server_address="api.openai.com",
@@ -425,7 +425,7 @@ def test_remote_agent_hook_called_on_fail(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        invocation = handler.start_invoke_remote_agent("openai")
+        invocation = handler.invoke_remote_agent("openai")
         invocation.fail(RuntimeError("remote agent crashed"))
 
         hook.on_completion.assert_called_once()
@@ -436,8 +436,8 @@ def test_agent_hook_called_with_empty_messages_when_none_set(self):
         hook = MagicMock()
         handler = self._make_handler(hook)
 
-        handler.start_invoke_local_agent("openai").stop()
-        handler.start_invoke_remote_agent("openai").stop()
+        handler.invoke_local_agent("openai").stop()
+        handler.invoke_remote_agent("openai").stop()
 
         for call in hook.on_completion.call_args_list:
             self.assertEqual(call.kwargs["inputs"], [])
@@ -448,5 +448,5 @@ def test_agent_hook_called_with_empty_messages_when_none_set(self):
     def test_agent_hook_not_called_when_not_set(self):
         # No hook — stop should not raise
         handler = self._make_handler()
-        handler.start_invoke_local_agent("openai").stop()
-        handler.start_invoke_remote_agent("openai").stop()
+        handler.invoke_local_agent("openai").stop()
+        handler.invoke_remote_agent("openai").stop()
diff --git a/util/opentelemetry-util-genai/tests/test_handler_metrics.py b/util/opentelemetry-util-genai/tests/test_handler_metrics.py
index 5e2a58fd..37a13cdd 100644
--- a/util/opentelemetry-util-genai/tests/test_handler_metrics.py
+++ b/util/opentelemetry-util-genai/tests/test_handler_metrics.py
@@ -27,7 +27,7 @@ def test_stop_llm_records_duration_and_tokens(self) -> None:
         )
         # Patch default_timer during start to ensure monotonic_start_s
         with patch("timeit.default_timer", return_value=1000.0):
-            invocation = handler.start_inference("prov", request_model="model")
+            invocation = handler.inference("prov", request_model="model")
         invocation.input_tokens = 5
         invocation.output_tokens = 7
 
@@ -82,7 +82,7 @@ def test_stop_llm_records_duration_and_tokens_with_additional_attributes(
             meter_provider=self.meter_provider,
         )
 
-        invocation = handler.start_inference(
+        invocation = handler.inference(
             "prov",
             request_model="model",
             server_address="custom.server.com",
@@ -121,7 +121,7 @@ def test_fail_llm_records_error_and_available_tokens(self) -> None:
         )
         # Patch default_timer during start to ensure monotonic_start_s
         with patch("timeit.default_timer", return_value=2000.0):
-            invocation = handler.start_inference("", request_model="err-model")
+            invocation = handler.inference("", request_model="err-model")
         invocation.input_tokens = 11
 
         error = Error(message="boom", type=ValueError)
@@ -193,7 +193,7 @@ def test_stop_embedding_records_duration_and_tokens(self) -> None:
         )
         # Patch default_timer during start to ensure monotonic_start_s
         with patch("timeit.default_timer", return_value=1000.0):
-            invocation = handler.start_embedding(
+            invocation = handler.embedding(
                 "embed-prov", request_model="embed-model"
             )
         invocation.input_tokens = 100
@@ -242,7 +242,7 @@ def test_stop_embedding_records_duration_with_additional_attributes(
             tracer_provider=self.tracer_provider,
             meter_provider=self.meter_provider,
         )
-        invocation = handler.start_embedding(
+        invocation = handler.embedding(
             "embed-prov",
             request_model="embed-model",
             server_address="embed.server.com",
@@ -279,7 +279,7 @@ def test_fail_embedding_records_error_and_duration(self) -> None:
             meter_provider=self.meter_provider,
         )
         with patch("timeit.default_timer", return_value=3000.0):
-            invocation = handler.start_embedding(
+            invocation = handler.embedding(
                 "embed-prov", request_model="embed-err-model"
             )
 
@@ -313,7 +313,7 @@ def test_stop_embedding_without_tokens(self) -> None:
             tracer_provider=self.tracer_provider,
             meter_provider=self.meter_provider,
         )
-        invocation = handler.start_embedding(
+        invocation = handler.embedding(
             "embed-prov", request_model="embed-model"
         )
         # input_tokens is not set
@@ -343,7 +343,7 @@ def test_stop_tool_records_duration(self) -> None:
             meter_provider=self.meter_provider,
         )
         with patch("timeit.default_timer", return_value=1000.0):
-            invocation = handler.start_tool("get_weather")
+            invocation = handler.tool("get_weather")
         invocation.metric_attributes = {"custom.key": "custom_value"}
 
         with patch("timeit.default_timer", return_value=1002.5):
@@ -371,7 +371,7 @@ def test_fail_tool_records_duration_with_error(self) -> None:
             meter_provider=self.meter_provider,
         )
         with patch("timeit.default_timer", return_value=500.0):
-            invocation = handler.start_tool("failing_tool")
+            invocation = handler.tool("failing_tool")
 
         error = Error(message="Tool execution failed", type=RuntimeError)
         with patch("timeit.default_timer", return_value=501.5):
diff --git a/util/opentelemetry-util-genai/tests/test_handler_workflow.py b/util/opentelemetry-util-genai/tests/test_handler_workflow.py
index f7fbb863..a304f5b8 100644
--- a/util/opentelemetry-util-genai/tests/test_handler_workflow.py
+++ b/util/opentelemetry-util-genai/tests/test_handler_workflow.py
@@ -52,12 +52,12 @@ class TelemetryHandlerWorkflowTest(_WorkflowTestBase):
     # ------------------------------------------------------------------
 
     def test_start_workflow_creates_span(self) -> None:
-        invocation = self.handler.start_workflow(name="my_workflow")
+        invocation = self.handler.workflow(name="my_workflow")
         self.assertIsNot(invocation.span, INVALID_SPAN)
         invocation.stop()
 
     def test_start_workflow_span_name(self) -> None:
-        invocation = self.handler.start_workflow(name="my_pipeline")
+        invocation = self.handler.workflow(name="my_pipeline")
         invocation.stop()
 
         spans = self._get_finished_spans()
@@ -65,7 +65,7 @@ def test_start_workflow_span_name(self) -> None:
         self.assertEqual(spans[0].name, "invoke_workflow my_pipeline")
 
     def test_start_workflow_span_name_without_name(self) -> None:
-        invocation = self.handler.start_workflow(name=None)
+        invocation = self.handler.workflow(name=None)
         invocation.stop()
 
         spans = self._get_finished_spans()
@@ -73,7 +73,7 @@ def test_start_workflow_span_name_without_name(self) -> None:
         self.assertEqual(spans[0].name, "invoke_workflow")
 
     def test_start_workflow_span_kind_is_internal(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.stop()
 
         spans = self._get_finished_spans()
@@ -82,7 +82,7 @@ def test_start_workflow_span_kind_is_internal(self) -> None:
 
     def test_start_workflow_records_monotonic_start(self) -> None:
         with patch("timeit.default_timer", return_value=500.0):
-            invocation = self.handler.start_workflow(name="wf")
+            invocation = self.handler.workflow(name="wf")
         self.assertEqual(invocation._monotonic_start_s, 500.0)
         invocation.stop()
 
@@ -91,14 +91,14 @@ def test_start_workflow_records_monotonic_start(self) -> None:
     # ------------------------------------------------------------------
 
     def test_stop_workflow_ends_span(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.stop()
 
         spans = self._get_finished_spans()
         self.assertEqual(len(spans), 1)
 
     def test_stop_workflow_sets_operation_name_attribute(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.stop()
 
         spans = self._get_finished_spans()
@@ -108,7 +108,7 @@ def test_stop_workflow_sets_operation_name_attribute(self) -> None:
         )
 
     def test_stop_workflow_sets_custom_attributes(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.attributes["custom.key"] = "custom_value"
         invocation.stop()
 
@@ -116,7 +116,7 @@ def test_stop_workflow_sets_custom_attributes(self) -> None:
         self.assertEqual(spans[0].attributes["custom.key"], "custom_value")
 
     def test_stop_workflow_returns_invocation(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.stop()
         spans = self._get_finished_spans()
         self.assertEqual(len(spans), 1)
@@ -126,7 +126,7 @@ def test_stop_workflow_returns_invocation(self) -> None:
     # ------------------------------------------------------------------
 
     def test_fail_workflow_sets_error_status(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         error = Error(message="something broke", type=RuntimeError)
         invocation.fail(error)
 
@@ -136,7 +136,7 @@ def test_fail_workflow_sets_error_status(self) -> None:
         self.assertEqual(spans[0].status.description, "something broke")
 
     def test_fail_workflow_sets_error_type_attribute(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         error = Error(message="bad", type=ValueError)
         invocation.fail(error)
 
@@ -144,7 +144,7 @@ def test_fail_workflow_sets_error_type_attribute(self) -> None:
         self.assertEqual(spans[0].attributes["error.type"], "ValueError")
 
     def test_fail_workflow_sets_operation_name_attribute(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         error = Error(message="fail", type=TypeError)
         invocation.fail(error)
 
@@ -155,7 +155,7 @@ def test_fail_workflow_sets_operation_name_attribute(self) -> None:
         )
 
     def test_fail_workflow_ends_span(self) -> None:
-        invocation = self.handler.start_workflow(name="wf")
+        invocation = self.handler.workflow(name="wf")
         invocation.fail(Error(message="err", type=RuntimeError))
         spans = self._get_finished_spans()
         self.assertEqual(len(spans), 1)
@@ -211,7 +211,7 @@ def get_description(self):
         )
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_workflow(name="my-workflow")
+        invocation = handler.workflow(name="my-workflow")
         invocation.stop()
 
         self.assertEqual(
diff --git a/util/opentelemetry-util-genai/tests/test_toolcall.py b/util/opentelemetry-util-genai/tests/test_toolcall.py
index 54648f06..0f65e441 100644
--- a/util/opentelemetry-util-genai/tests/test_toolcall.py
+++ b/util/opentelemetry-util-genai/tests/test_toolcall.py
@@ -40,7 +40,7 @@ def test_toolcallrequest_is_message_part():
 def test_toolcall_inherits_from_genaiinvocation():
     """ToolInvocation inherits from GenAIInvocation for lifecycle management"""
     handler = _make_handler()
-    tc = handler.start_tool("get_weather", arguments={"city": "Paris"})
+    tc = handler.tool("get_weather", arguments={"city": "Paris"})
     assert isinstance(tc, GenAIInvocation)
     assert not isinstance(tc, ToolCallRequest)
     tc.stop()
@@ -49,7 +49,7 @@ def test_toolcall_inherits_from_genaiinvocation():
 def test_toolcall_has_attributes_dict():
     """ToolInvocation inherits attributes dict from GenAIInvocation"""
     handler = _make_handler()
-    tc = handler.start_tool("test")
+    tc = handler.tool("test")
     tc.attributes["custom.key"] = "value"
     assert tc.attributes["custom.key"] == "value"
     tc.stop()
@@ -69,7 +69,7 @@ def test_toolcallrequest_in_message_part_union():
 def test_toolcall_operation_name():
     """ToolInvocation operation_name is fixed to execute_tool"""
     handler = _make_handler()
-    tc = handler.start_tool("my_tool")
+    tc = handler.tool("my_tool")
     assert tc._operation_name == "execute_tool"
     tc.stop()
 
@@ -160,7 +160,7 @@ def get_description(self):
     sampler_provider.add_span_processor(SimpleSpanProcessor(span_exporter))
     handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-    invocation = handler.start_tool(
+    invocation = handler.tool(
         "get_weather",
         tool_call_id="call_123",
         tool_type="function",
diff --git a/util/opentelemetry-util-genai/tests/test_utils.py b/util/opentelemetry-util-genai/tests/test_utils.py
index 969dc965..4059bc0c 100644
--- a/util/opentelemetry-util-genai/tests/test_utils.py
+++ b/util/opentelemetry-util-genai/tests/test_utils.py
@@ -320,7 +320,7 @@ def test_llm_manual_start_and_stop_creates_span(self):
         message = _create_input_message("hi")
         chat_generation = _create_output_message("ok")
 
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="manual-model"
         )
         invocation.input_messages = [message]
@@ -380,7 +380,7 @@ def get_description(self):
 
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_inference(
+        invocation = handler.inference(
             "test-provider",
             request_model="sampler-model",
             server_address="api.example.com",
@@ -437,13 +437,13 @@ def get_description(self):
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
         # This invocation should be dropped
-        invocation = handler.start_inference(
+        invocation = handler.inference(
             "test-provider", request_model="rejected-model"
         )
         invocation.stop()
 
         # This invocation should be recorded
-        invocation = handler.start_inference(
+        invocation = handler.inference(
             "test-provider", request_model="accepted-model"
         )
         invocation.stop()
@@ -478,7 +478,7 @@ def get_description(self):
         )
         handler = TelemetryHandler(tracer_provider=sampler_provider)
 
-        invocation = handler.start_embedding(
+        invocation = handler.embedding(
             "test-provider",
             request_model="embed-model",
             server_address="embed.example.com",
@@ -498,7 +498,7 @@ def get_description(self):
         assert captured_attributes[server_attributes.SERVER_PORT] == 443
 
     def test_llm_span_finish_reasons_without_output_messages(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="model-without-output"
         )
         invocation.finish_reasons = ["length"]
@@ -527,7 +527,7 @@ def test_llm_span_finish_reasons_without_output_messages(self):
         )
 
     def test_llm_span_finish_reasons_from_invocation(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="model-reasons"
         )
         invocation.finish_reasons = ["stop", "length", "stop"]
@@ -542,7 +542,7 @@ def test_llm_span_finish_reasons_from_invocation(self):
         )
 
     def test_llm_span_finish_reasons_from_output_messages(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="model-output-reasons"
         )
         assert invocation.span is not None
@@ -561,7 +561,7 @@ def test_llm_span_finish_reasons_from_output_messages(self):
         )
 
     def test_llm_span_uses_expected_schema_url(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "schema-provider", request_model="schema-model"
         )
         assert invocation.span is not None
@@ -584,7 +584,7 @@ def test_llm_span_uses_expected_schema_url(self):
         emit_event="true",
     )
     def test_llm_log_uses_expected_schema_url(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "schema-provider", request_model="schema-model"
         )
         invocation.output_messages = [_create_output_message()]
@@ -639,12 +639,12 @@ def test_parent_child_span_relationship(self):
         emit_event="",
     )
     def test_embedding_parent_child_span_relationship(self):
-        parent_invocation = self.telemetry_handler.start_embedding(
+        parent_invocation = self.telemetry_handler.embedding(
             "test-provider", request_model="embed-parent-model"
         )
         parent_invocation.input_tokens = 10
         assert parent_invocation.span is not None
-        child_invocation = self.telemetry_handler.start_embedding(
+        child_invocation = self.telemetry_handler.embedding(
             "test-provider", request_model="embed-child-model"
         )
         child_invocation.input_tokens = 5
@@ -679,7 +679,7 @@ def test_llm_parent_embedding_child_span_relationship(self):
             "test-provider", request_model="parent-model"
         ) as parent_invocation:
             parent_invocation.input_messages = [message]
-            child_invocation = self.telemetry_handler.start_embedding(
+            child_invocation = self.telemetry_handler.embedding(
                 "test-provider", request_model="embed-child-model"
             )
             child_invocation.input_tokens = 3
@@ -790,7 +790,7 @@ class BoomError(RuntimeError):
         emit_event="",
     )
     def test_embedding_manual_start_and_stop_creates_span(self):
-        invocation = self.telemetry_handler.start_embedding(
+        invocation = self.telemetry_handler.embedding(
             "test-provider",
             request_model="embed-model",
             server_address="custom.server.com",
@@ -831,7 +831,7 @@ def test_fail_with_exception_sets_error_status_and_type(self):
         class BoomError(RuntimeError):
             pass
 
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="test-model"
         )
         invocation.fail(BoomError("boom"))
diff --git a/util/opentelemetry-util-genai/tests/test_utils_events.py b/util/opentelemetry-util-genai/tests/test_utils_events.py
index bbdd342a..6c5686d0 100644
--- a/util/opentelemetry-util-genai/tests/test_utils_events.py
+++ b/util/opentelemetry-util-genai/tests/test_utils_events.py
@@ -61,7 +61,7 @@ def tearDown(self):
         emit_event="true",
     )
     def test_emits_llm_event(self):
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="event-model"
         )
         invocation.input_messages = [_create_input_message("test query")]
@@ -143,7 +143,7 @@ def test_emits_llm_event_and_span(self):
         chat_generation = _create_output_message("combined response")
         system_instruction = _create_system_instruction("System prompt here")
 
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="combined-model"
         )
         invocation.input_messages = [message]
@@ -200,7 +200,7 @@ class TestError(RuntimeError):
             pass
 
         message = _create_input_message("error test")
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="error-model"
         )
         invocation.input_messages = [message]
@@ -236,7 +236,7 @@ def test_does_not_emit_llm_event_when_emit_event_false(self):
         message = _create_input_message("emit false test")
         chat_generation = _create_output_message("emit false response")
 
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="emit-false-model"
         )
         invocation.input_messages = [message]
@@ -254,7 +254,7 @@ def test_does_not_emit_llm_event_when_emit_event_false(self):
     )
     def test_does_not_emit_llm_event_by_default_for_no_content(self):
         """Test that event is not emitted by default when content_capturing is NO_CONTENT and OTEL_INSTRUMENTATION_GENAI_EMIT_EVENT is not set."""
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="default-model"
         )
         invocation.input_messages = [_create_input_message("default test")]
@@ -274,7 +274,7 @@ def test_does_not_emit_llm_event_by_default_for_no_content(self):
     )
     def test_does_not_emit_llm_event_by_default_for_span_only(self):
         """Test that event is not emitted by default when content_capturing is SPAN_ONLY and OTEL_INSTRUMENTATION_GENAI_EMIT_EVENT is not set."""
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="default-model"
         )
         invocation.input_messages = [_create_input_message("default test")]
@@ -294,7 +294,7 @@ def test_does_not_emit_llm_event_by_default_for_span_only(self):
     )
     def test_emits_llm_event_by_default_for_event_only(self):
         """Test that event is emitted by default when content_capturing is EVENT_ONLY and OTEL_INSTRUMENTATION_GENAI_EMIT_EVENT is not set."""
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="default-model"
         )
         invocation.input_messages = [_create_input_message("default test")]
@@ -322,7 +322,7 @@ def test_emits_llm_event_by_default_for_span_and_event(self):
         chat_generation = _create_output_message("span and event response")
         system_instruction = _create_system_instruction("System prompt")
 
-        invocation = self.telemetry_handler.start_inference(
+        invocation = self.telemetry_handler.inference(
             "test-provider", request_model="span-and-event-model"
         )
         invocation.input_messages = [message]
diff --git a/util/opentelemetry-util-genai/tests/test_workflow_invocation.py b/util/opentelemetry-util-genai/tests/test_workflow_invocation.py
index e6959ed5..5a8a559d 100644
--- a/util/opentelemetry-util-genai/tests/test_workflow_invocation.py
+++ b/util/opentelemetry-util-genai/tests/test_workflow_invocation.py
@@ -27,7 +27,7 @@ def setUp(self):
         self.handler = TelemetryHandler(tracer_provider=tracer_provider)
 
     def test_default_values(self):
-        invocation = self.handler.start_workflow(name=None)
+        invocation = self.handler.workflow(name=None)
         invocation.stop()
         assert invocation.name is None
         assert invocation._operation_name == "invoke_workflow"
@@ -37,15 +37,13 @@ def test_default_values(self):
         assert not invocation.attributes
 
     def test_custom_name(self):
-        invocation = self.handler.start_workflow(
-            name="customer_support_pipeline"
-        )
+        invocation = self.handler.workflow(name="customer_support_pipeline")
         invocation.stop()
         assert invocation.name == "customer_support_pipeline"
 
     def test_with_input_messages(self):
         msg = InputMessage(role="user", parts=[Text(content="hello")])
-        invocation = self.handler.start_workflow(name="test")
+        invocation = self.handler.workflow(name="test")
         invocation.input_messages = [msg]
         invocation.stop()
         assert len(invocation.input_messages) == 1
@@ -55,14 +53,14 @@ def test_with_output_messages(self):
         msg = OutputMessage(
             role="assistant", parts=[Text(content="hi")], finish_reason="stop"
         )
-        invocation = self.handler.start_workflow(name="test")
+        invocation = self.handler.workflow(name="test")
         invocation.output_messages = [msg]
         invocation.stop()
         assert len(invocation.output_messages) == 1
         assert invocation.output_messages[0].finish_reason == "stop"
 
     def test_inherits_genai_invocation(self):
-        invocation = self.handler.start_workflow(name="test")
+        invocation = self.handler.workflow(name="test")
         invocation.attributes["key"] = "value"
         invocation.stop()
         spans = self.span_exporter.get_finished_spans()
@@ -71,16 +69,16 @@ def test_inherits_genai_invocation(self):
 
     def test_default_lists_are_independent(self):
         """Ensure separate invocations get separate list instances."""
-        inv1 = self.handler.start_workflow(name=None)
-        inv2 = self.handler.start_workflow(name=None)
+        inv1 = self.handler.workflow(name=None)
+        inv2 = self.handler.workflow(name=None)
         inv1.input_messages.append(InputMessage(role="user", parts=[]))
         assert len(inv2.input_messages) == 0
         inv1.stop()
         inv2.stop()
 
     def test_default_attributes_are_independent(self):
-        inv1 = self.handler.start_workflow(name=None)
-        inv2 = self.handler.start_workflow(name=None)
+        inv1 = self.handler.workflow(name=None)
+        inv2 = self.handler.workflow(name=None)
         inv1.attributes["foo"] = "bar"
         assert "foo" not in inv2.attributes
         inv1.stop()
@@ -93,7 +91,7 @@ def test_full_construction(self):
             parts=[Text(content="answer")],
             finish_reason="stop",
         )
-        invocation = self.handler.start_workflow(name="my_workflow")
+        invocation = self.handler.workflow(name="my_workflow")
         invocation.input_messages = [inp]
         invocation.output_messages = [out]
         invocation.stop()