- Extract fetch_neo4j_assistant_aliases() into deduped_and_disamb.py as
single source of truth, replacing inline Cypher in write_tools and
extraction_orchestrator
- Normalize USER_PLACEHOLDER_NAMES to lowercase and apply .lower() on
all comparisons to prevent case-variant names leaking into aliases
- Extract user aliases from raw dialog statements instead of post-dedup
entities to bypass merge pollution
- Add alias cross-cleaning step in _normalize_special_entity_names to
strip AI assistant aliases from user entities before dedup
- Call clean_cross_role_aliases after second-layer dedup to handle
historical dirty data merged from Neo4j
- Fix syntax error in prompt_utils.py (ontology_types variable assignment)
- Trigger opening statement on new conversation in run/run_stream
- Fix /opening endpoint to support workflow app type
- Fix features field missing in workflow config release snapshot
- Knowledge node returns citations alongside chunks
- Aggregate citations from all knowledge nodes in result builder
- Filter citations based on features.citation.enabled switch
- Fix WorkflowConfigCreate circular import in app_schema
- Add speaker context to triplet extraction prompt to distinguish alias ownership
- Add explicit examples and rules in extract_triplet.jinja2 for user vs AI alias attribution
- Introduce cross-role merge protection in dedup (accurate, fuzzy, and LLM stages)
- Normalize special entity names (用户/AI助手) before deduplication
- Add clean_cross_role_aliases() to sanitize aliases before Neo4j write
- Refactor _update_end_user_other_name to merge aliases from PgSQL instead of Neo4j
- Filter AI assistant aliases from user alias extraction in orchestrator
- Remove unused TaskStatusResponse import from memory_api_schema
- Add _sanitize_task_result() helper function to convert non-serializable types (UUID, datetime) to strings
- Update get_write_task_status endpoint to use sanitization instead of TaskStatusResponse validation
- Update get_read_task_status endpoint to use sanitization instead of TaskStatusResponse validation
- Ensures Celery task results are properly JSON-serializable before returning to clients
- Rename endpoints from write_api_service/read_api_service to write/read for clarity
- Add async task-based endpoints (/write, /read) that dispatch to Celery with fair locking
- Add task status polling endpoints (/write/status, /read/status) to check async operation results
- Add synchronous endpoints (/write/sync, /read/sync) for blocking operations with direct results
- Introduce TaskStatusResponse schema for task status polling responses
- Add MemoryWriteSyncResponse and MemoryReadSyncResponse schemas for sync operations
- Implement write_memory_sync and read_memory_sync methods in MemoryAPIService
- Remove await from async service calls in task-based endpoints (now handled by Celery)
- Add Query parameter import for task_id in status endpoints
- Update docstrings to clarify async vs sync behavior and task polling workflow
- Integrate task_service for retrieving Celery task results
- Increased max_connections for Redis pool from 10 to 100.
- Extended socket_timeout from 5 to 10 seconds.
- Added retry mechanism with exponential backoff for Redis operations in `RedisFairLock`.
- Create new memory_config_api_controller.py for dedicated memory configuration management
- Add /end_user/info GET endpoint to retrieve end user information (aliases, metadata)
- Add /end_user/info/update POST endpoint to update end user details
- Move /memory/configs endpoint from memory_api_controller to memory_config_api_controller
- Extract _get_current_user helper function to build user context from API key auth
- Support optional app_id parameter in end user creation with UUID validation
- Update service controller imports with alphabetical ordering and multi-line formatting
- Register memory_config_api_controller router in service module initialization
- Refactor memory_api_controller imports for consistency and clarity
- Removed try-except block for importing `audit_logger` and directly imported it.
- Removed redundant checks for `audit_logger` being `None` before logging operations.
- Added a check in `MemoryPerceptualService` to return `None` if `model_config` or `llm` is `None`.
