- Remove dead elif branch in retrieve_api_base derivation
- Replace unreachable try/except httpx.HTTPStatusError around GET
calls with logging inside the status_code check (HTTPHandler.get()
does not call raise_for_status())
- Add comments noting HTTPHandler.get()/AsyncHTTPHandler.get() do not
accept a timeout parameter
Made-with: Cursor
- Document intentional legacy-matching behavior: when user_id is
provided to an org admin, no org filter is applied (returns all of
that user's teams across all orgs, same as legacy endpoint)
- Fix two existing security tests to properly patch user_api_key_cache,
proxy_logging_obj, and get_user_object instead of relying on
incidental error handling
- Add three new org admin test cases:
- Org admin sees org-scoped teams (200 with correct where clause)
- Org admin rejected when filtering by other org (403)
- Org admin with user_id filter returns target user's teams
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Pydantic model instances (ResponseFunctionToolCall, etc.) from first_response.output
were included raw in follow_up_input; the transformation layer expects plain dicts and
called .get() on them, raising AttributeError. Serialize via model_dump(exclude_none=True).
Made-with: Cursor
- Remove dead `should_use_emulated_file_search` (main.py uses its own inline guard)
- Remove dead `fallback_vector_store_ids` param from `_run_vector_searches`
- Include all first_response.output items in follow_up_input so text blocks/reasoning
from providers like Anthropic aren't dropped from conversation context
- Accumulate first provider call's response_cost into synthesized _hidden_params so
billing callbacks see the total cost of both emulated-flow LLM calls
- Remove broad tools=[] filter from transformation.py (backward-incompatible); the
follow-up call already passes tools=None which is filtered by the v is not None guard
Made-with: Cursor
- Replace raw find_unique with get_user_object in
_build_team_list_where_conditions for cache/metrics consistency
- Remove over-complex OR clause for org admin + user_id: when user_id
is provided, filter by that user's direct team memberships (same as
regular users) since the access control gate already verified the
org admin's authority
- Preserve caller-supplied organization_id instead of overwriting with
org_admin_org_ids
- Update test mock to match get_user_object call path
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- _get_org_admin_org_ids: catch only ValueError (user not found) instead
of bare Exception — DB errors now propagate as 500s instead of silently
demoting org admins to regular users
- _build_team_list_where_conditions: return None (not a sentinel string)
when user has no team memberships; list_team_v2 short-circuits to empty
response without hitting the DB
- Org admin + team_id + user_id: use exact team_id match with org scope
instead of OR expansion that effectively ignored the team_id filter
- Org admin + user_id (no team_id): OR(org teams, direct memberships)
now matches legacy _authorize_and_filter_teams behaviour
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Replace misleading endpoint extraction with explicit endpoint = "cancel"
- Compute retrieve_api_base from URL components directly instead of
stripping ":cancel" from the post-proxy URL, removing the hard ValueError
that broke any custom Vertex AI proxy configuration
- Align cancel_batch provider priority in proxy endpoints to match
create_batch order: body field → request headers → query params → default
Made-with: Cursor
- Fix org admin own-query regression: always check org admin status
before the standard route check so own-queries see all org teams
- Clear user_id when org admin is detected so org scope replaces
user-membership scope
- Remove dead isinstance(organization_id, list) branch
- Remove unused datetime import
- Remove orphaned _convert_teams_to_response helper
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add _is_reasoning_effort_level_explicitly_disabled to use opt-out semantics
for minimal effort: unknown/unlisted models pass through, only blocked when
the model map explicitly sets supports_minimal_reasoning_effort=false.
xhigh keeps opt-in semantics (must be explicitly supported).
Adds test for unknown-model passthrough and explicit-disabled detection.
Made-with: Cursor
- Add try/except httpx.HTTPStatusError blocks in _async_cancel_batch for
both POST cancel and GET retrieve calls, with verbose_logger error logging
- Fix endpoint extraction inconsistency: compute endpoint from URL without
:cancel suffix so it matches behaviour of create_batch/retrieve_batch
- Add explicit validation that api_base ends with ':cancel' before
stripping it, raising a descriptive error for unsupported custom proxy
URL rewriting scenarios
- Use string-based patch() in test instead of patch.object() for robustness
against import order changes
Made-with: Cursor
check_and_create_cache calls check_cache first (to avoid duplicates),
which also needs model for the URL when api_base is set. Without this,
the full flow still raises ValueError before reaching the create step.
Add org admin support to /v2/team/list so org admins can list teams
within their organizations instead of getting 401. Also enrich the
response with members_count and add missing indexes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Fix bug where only last text item's annotations were preserved when
thread.message.completed contained multiple text content items
- Accumulate annotations via extend() instead of overwriting
- Add test_azure_ai_agents_streaming_annotations_from_completed_message
- Add test_azure_ai_agents_streaming_accumulates_annotations_from_multiple_text_items
Addresses Greptile review on PR #23849
Made-with: Cursor
_get_token_and_url_context_caching() was hardcoding model=None when
calling _check_custom_proxy(), which raises ValueError when api_base
is set because Gemini proxy URLs need the model name:
{api_base}/models/{model}:cachedContents
Fixes#23846
Azure GPT-5.4+ models now get the same auto-routing treatment as OpenAI
when both `reasoning_effort` and `tools` are used in `litellm.completion()`.
Previously, `reasoning_effort` was silently dropped for Azure; now the
request is bridged to the Responses API which supports both parameters.
Fixes#23914
Fixes#23890 — Mistral's Voxtral transcription with `diarize=true` returns
`segments` (with speaker_id, timestamps) and `language`, but these fields
were dropped when mapping the response to TranscriptionResponse.
Address review feedback from greptile — use new_callable=AsyncMock
on the concurrent test's patch.object to ensure the mock is properly
typed as async, even though side_effect already handles the coroutine.
The release job was failing with "Resource not accessible by integration"
because other jobs explicitly set permissions, causing GitHub to scope the
default token down for all jobs. The release job needs contents:write to
create GitHub releases.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
