…error

fix: upgraded mcp sdk depency version for fixing ClosedResourceError

…il traces

- Remove redundant set_attributes() call from raw_gen_ai_request spans so they
  only contain provider-specific llm.{provider}.{param} attributes (Issue #3)
- Replace full set_attributes() on litellm_proxy_request with minimal attributes
  (model, call_type, response.id) when litellm_request child span exists (Issue #4)
- Skip guardrail span creation when no parent context exists to prevent orphaned
  root-level traces in the "All Spans" view (Issue #5)
- Fall back to litellm_call_id for gen_ai.response.id when response object has no
  id field (EmbeddingResponse, ImageResponse) (Issue BerriAI#8)

Co-Authored-By: Claude Opus 4.6 <[email protected]>

…onse IDs

Addresses 4 critical OpenTelemetry span issues in LiteLLM:

Issue #3: Remove redundant attributes from raw_gen_ai_request spans
- Removed self.set_attributes() call that was duplicating all parent span
  attributes (gen_ai.*, metadata.*) onto the raw span
- Raw span now only contains provider-specific llm.{provider}.* attributes
- Reduces storage and eliminates search confusion from duplicate data

Issue #4: Prevent attribute duplication on litellm_proxy_request parent span
- When litellm_request child span exists, removed redundant
  set_attributes() call on the parent proxy span
- Child span already carries all attributes; parent duplication doubles
  storage and complicates search

Issue #5: Fix orphaned guardrail traces
- Guardrail spans were created with context=None when no parent proxy span
  existed, resulting in orphaned root spans (separate trace_id)
- Added _resolve_guardrail_context() helper to ensure guardrails always
  have a valid parent (litellm_request or proxy span)
- Applied fix to both _handle_success and _handle_failure paths

Issue BerriAI#8: Add gen_ai.response.id for embeddings and image generation
- EmbeddingResponse and ImageResponse types don't have provider response IDs
- Added fallback to standard_logging_payload["id"] (litellm call ID) for
  correlation across LiteLLM UI, Phoenix traces, and provider logs
- Completions still use provider ID (e.g. "chatcmpl-xxx") when available

Tests added:
- TestRawSpanAttributeIsolation: Verify raw span has no gen_ai/metadata attrs
- TestNoParentSpanDuplication: Verify parent span doesn't get duplicated attrs
- TestGuardrailSpanParenting: Verify guardrails are children (not orphaned)
- TestResponseIdFallback: Verify response ID set for all call types

All existing OTEL tests pass (73 passed, 14 pre-existing protocol failures).

Co-Authored-By: Claude Haiku 4.5 <[email protected]>

docs: Learn page updates, card links, integrations, sidebar changes

- Gap #1: block key/update on jwt-bound keys for non-admins (403)
- Gap #2: stamp jwt_bound metadata + restrict allowed_routes on mapping creation
- Gap #3: /jwt_client/new unified endpoint — atomically creates key + mapping, returns cleartext key
- Gap #4: unregistered_jwt_client_behavior config (reject/fallback_team_mapping/auto_register)
- Gap #5: issuer column on LiteLLM_JWTKeyMapping for multi-IdP support
- Gap #6: /jwt/key/mapping/info and /jwt_client/update expose virtual key fields
- 31 unit tests covering all gaps

test(): add safeguard tests

…riAI#6 BerriAI#7

- Move MAVVRIK_EXPORT_INTERVAL_MINUTES/JOB_NAME to top-level imports (#3)
- Extract _schedule_job() helper to eliminate 3× duplicated scheduler
  registration blocks in initialize(), update_settings(), proxy_server.py (BerriAI#4)
- delete() now logs informatively on LookupError instead of bare pass (BerriAI#5)
- Logger.__init__ warns at startup if credentials are not configured (BerriAI#6)
- Rename _stream_pages/_stream_upload to stream_pages/stream_upload (BerriAI#7)

Co-Authored-By: Claude Sonnet 4.6 (1M context) <[email protected]>

…-tab links

Compute a safeHref that only passes through http/https URLs to prevent
DOM-based XSS via tainted window.location data in href sinks (items BerriAI#4 and BerriAI#5).

Co-Authored-By: Claude Sonnet 4.6 <[email protected]>

…source_error

fix: upgraded mcp sdk depency version for fixing ClosedResourceError

…onse IDs

Addresses 4 critical OpenTelemetry span issues in LiteLLM:

Issue BerriAI#3: Remove redundant attributes from raw_gen_ai_request spans
- Removed self.set_attributes() call that was duplicating all parent span
  attributes (gen_ai.*, metadata.*) onto the raw span
- Raw span now only contains provider-specific llm.{provider}.* attributes
- Reduces storage and eliminates search confusion from duplicate data

Issue BerriAI#4: Prevent attribute duplication on litellm_proxy_request parent span
- When litellm_request child span exists, removed redundant
  set_attributes() call on the parent proxy span
- Child span already carries all attributes; parent duplication doubles
  storage and complicates search

Issue BerriAI#5: Fix orphaned guardrail traces
- Guardrail spans were created with context=None when no parent proxy span
  existed, resulting in orphaned root spans (separate trace_id)
- Added _resolve_guardrail_context() helper to ensure guardrails always
  have a valid parent (litellm_request or proxy span)
- Applied fix to both _handle_success and _handle_failure paths

Issue BerriAI#8: Add gen_ai.response.id for embeddings and image generation
- EmbeddingResponse and ImageResponse types don't have provider response IDs
- Added fallback to standard_logging_payload["id"] (litellm call ID) for
  correlation across LiteLLM UI, Phoenix traces, and provider logs
- Completions still use provider ID (e.g. "chatcmpl-xxx") when available

Tests added:
- TestRawSpanAttributeIsolation: Verify raw span has no gen_ai/metadata attrs
- TestNoParentSpanDuplication: Verify parent span doesn't get duplicated attrs
- TestGuardrailSpanParenting: Verify guardrails are children (not orphaned)
- TestResponseIdFallback: Verify response ID set for all call types

All existing OTEL tests pass (73 passed, 14 pre-existing protocol failures).

docs: Learn page updates, card links, integrations, sidebar changes