refactor: clear fresh tech debt from the last 24 hours (2026-09-10)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

litellm/litellm_core_utils/prompt_templates/common_utils.py

@@ -2014,11 +2014,11 @@ def strip_encrypted_reasoning_from_messages(messages: object) -> None:
     """
     if not isinstance(messages, list):
         return
-    for content in _anthropic_content_lists(cast(list[object], messages)):  # cast-ok: untyped client json
+    for content in anthropic_content_lists(cast(list[object], messages)):  # cast-ok: untyped client json
         _strip_encrypted_reasoning_from_blocks(content)
 
 
-def _anthropic_content_lists(messages: Sequence[object]) -> Iterator[object]:
+def anthropic_content_lists(messages: Sequence[object]) -> Iterator[object]:
     return (
         cast(list[object], content)  # cast-ok: narrowed by isinstance
         for message in messages

litellm/litellm_core_utils/streaming_handler.py

@@ -1480,7 +1480,7 @@ class CustomStreamWrapper:
                 "is_finished": chunk_finish_reason is not None,
                 "finish_reason": chunk_finish_reason,
                 "original_chunk": cached_chunk,
-                "tool_calls": (getattr(cached_choice.delta, "tool_calls", None) if cached_choice is not None else None),
+                "tool_calls": cached_choice.delta.tool_calls if cached_choice is not None else None,
             }
 
             completion_obj["content"] = response_obj["text"]

litellm/llms/anthropic/experimental_pass_through/responses_adapters/transformation.py

@@ -171,7 +171,7 @@ class LiteLLMAnthropicToResponsesAPIAdapter:
         cls,
         summary: Iterable[object],
         encrypted_content: object,
-    ) -> dict[str, Any] | None:  # mutable-ok: API message payload
+    ) -> dict[str, object] | None:  # mutable-ok: API message payload
         """The one Anthropic block for a Responses reasoning item.
 
         The item's encrypted reasoning rides the block's opaque field (`signature`, or
@@ -200,7 +200,7 @@ class LiteLLMAnthropicToResponsesAPIAdapter:
     @classmethod
     def _assistant_group_to_input_items(
         cls, group: tuple[Mapping[str, object], ...]
-    ) -> tuple[dict[str, Any], ...]:  # mutable-ok: API message payload
+    ) -> tuple[dict[str, object], ...]:  # mutable-ok: API message payload
         first: Final = group[0]
         btype: Final = first.get("type")
         if btype in ("thinking", "redacted_thinking"):

litellm/proxy/management_endpoints/cost_tracking_settings.py

@@ -634,7 +634,6 @@ async def estimate_cost(
     # Pinning one moment keeps an off-peak window that opens mid-quote from pricing the totals on
     # one side of it and the reported rates on the other.
     with pinned_billing_time(current_billing_time()):
-        # Use completion_cost which handles all the logic including margins/discounts
         try:
             cost_per_request: Final = completion_cost(
                 completion_response=mock_response,

litellm/responses/streaming_iterator.py

@@ -1323,7 +1323,7 @@ def _stamp_responses_usage_cost(
     if usage_obj is None:
         return
     response_obj.usage = usage_obj  # rebind-ok: the stamped cost has to ride on the response the client receives
-    if isinstance(getattr(usage_obj, "cost", None), (int, float)):
+    if isinstance(usage_obj.cost, (int, float)):
         return
     try:
         cost: Final[float | None] = logging_obj._response_cost_calculator(result=response_obj)

litellm/router_utils/pre_call_checks/encrypted_content_affinity_check.py

@@ -49,6 +49,7 @@ from litellm.exceptions import (
 )
 from litellm.integrations.custom_logger import CustomLogger, Span
 from litellm.litellm_core_utils.prompt_templates.common_utils import (
+    anthropic_content_lists,
     encrypted_content_of_block,
     strip_encrypted_reasoning_from_messages,
 )
@@ -154,10 +155,7 @@ class EncryptedContentAffinityCheck(CustomLogger):
             return iter(())
         return (
             cast(Mapping[str, object], block)  # cast-ok: narrowed by isinstance
-            for message in cast(list[object], messages)  # cast-ok: narrowed by isinstance
-            if isinstance(message, Mapping)
-            for content in (cast(Mapping[str, object], message).get("content"),)  # cast-ok: narrowed by isinstance
-            if isinstance(content, list)
+            for content in anthropic_content_lists(cast(list[object], messages))  # cast-ok: narrowed by isinstance
             for block in cast(list[object], content)  # cast-ok: narrowed by isinstance
             if isinstance(block, Mapping)
         )