* 去掉MCP框架，重构
* 去掉MCP框架，重构
* 去掉MCP框架，重构
* 去掉MCP框架，重构
* 去掉MCP框架，重构
* 去掉MCP框架，重构
* 去掉MCP框架，重构
* feat(celery): add comprehensive logging to worker and write task
- Initialize logging system in Celery worker entry point with LoggingConfig
- Add logger instance and startup message to celery_worker.py
- Reorganize imports in tasks.py for better readability and consistency
- Add detailed logging to write_message_task for debugging and monitoring
- Log task start with group_id, config_id, and storage_type parameters
- Log service execution and completion status with results
- Add exception handling with error logging and stack trace capture
- Log task completion time and Celery task ID for performance tracking
- Improves observability and troubleshooting of async task execution
* 去掉MCP框架，重构
* 去掉MCP框架，重构
---------
Co-authored-by: Ke Sun <kesun5@illinois.edu>
* [feature]Emotional memory cache
* [feature]Implicit memory cache
* [changes]Modify the expiration time of implicit memory to 24 hours.
* [feature]Emotional memory cache
* [feature]Implicit memory cache
* [changes]Modify the expiration time of implicit memory to 24 hours.
* [changes]Modify the code based on the AI review
* [feature]Emotional memory cache
* [feature]Implicit memory cache
* [changes]Modify the expiration time of implicit memory to 24 hours.
* [feature]Implicit memory cache
* [changes]Modify the code based on the AI review
- Remove template_service extraction and template rendering logic
- Remove LLM client initialization from MemoryClientFactory
- Remove structured response call to LLM with RetrieveSummaryResponse model
- Replace LLM-based answer generation with direct retrieval information
- Simplify response to use raw retrieved info or default fallback message
- Update logging to reflect non-LLM quick answer approach
- Reduce unnecessary dependencies and improve performance by eliminating LLM call overhead
- Add [PERF] prefixed logging throughout hybrid search pipeline for better performance visibility
- Break down latency metrics with separate timing for config loading, embedder initialization, and rerank computation
- Format latency breakdown as JSON in performance summary logs
- Optimize batch_record_access to process node access records in parallel using asyncio.gather instead of sequential processing
- Add performance timing instrumentation for forgetting config loading and rerank computation stages
- Reorganize imports in access_history_manager for consistency
- Improve observability of search performance bottlenecks through structured logging
* [fix]Fix the return of the "content" attribute
* [changes]Improve the code based on AI review
* Apply suggestion from @sourcery-ai[bot]
Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>
* [fix]Fix the return of the "content" attribute
* [changes]Improve the code based on AI review
* Apply suggestion from @sourcery-ai[bot]
Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>
* [changes]Improve the code based on AI review
---------
Co-authored-by: sourcery-ai[bot] <58596630+sourcery-ai[bot]@users.noreply.github.com>
* [refactor]Reconstructing forgotten, emotional, situational, and explicit memory statistics
* [refactor]Reconstructing forgotten, emotional, situational, and explicit memory statistics
* [changes]Improve the code based on AI review
* fix(workflow): use loose rendering for end-node variables
* fix(workflow): use int type for memory node config id
* fix(workflow): handle missing environment variable defaults
* fix(workflow): render jinja variables with actual values in non-strict mode
* fix(workflow): support reordering without a rerank model in knowledge base
* fix(workflow): fix typo in key value
