The form-data validation now ensures all items in the list are of type HttpFormData. Truncation length for embedding inputs is reduced from 8191 to 8000 to accommodate tokenizer differences and avoid overflow. Excel parsing now disables chunking by setting chunk_token_num to 0, aligning with intended behavior for structured file ingestion.
- Updated form-data handling to accept both single FileVariable and ArrayVariable containing FileVariable for file uploads
- Fixed HTTP client redirect handling by enabling follow_redirects=True when downloading remote files
- Adjusted config validation to correctly require list type for form-data fields instead of HttpFormData class
- Replace curl with urllib.request for API calls to improve portability
- Move API key to environment variable for better security practices
- Inline Python script using heredoc for cleaner workflow definition
- Add intermediate file (ai_summary.txt) to separate concerns between API call and output handling
- Simplify JSON payload construction using Python's json module
- Improve error handling with fallback message for failed AI generation
- Update prompt instruction to request numbered list format
- Remove title and preamble from AI output for cleaner formatting
- Improve clarity by specifying "要点" (key points) in prompt
- Enhance consistency of release notification messages
- Extract WeChat notification payload construction from inline curl command
- Move environment variables to explicit env section for clarity
- Build JSON payload using Python for better string handling and readability
- Write payload to temporary file and pass to curl via -d @wechat.json
- Improves maintainability and reduces shell string escaping complexity
- Remove build_wechat_payload.py script and consolidate payload construction directly in workflow
- Eliminate intermediate environment variables and file I/O operations for cleaner execution
- Inline AI summary payload generation into curl request
- Inline WeChat notification payload generation into curl request
- Remove unnecessary checkout step since script is no longer needed
- Simplify workflow by reducing file dependencies and improving readability
- Create new `.github/scripts/build_wechat_payload.py` to handle WeChat payload generation
- Replace inline Python string concatenation with dedicated script for better maintainability
- Add checkout step to access the script during workflow execution
- Simplify workflow by delegating payload construction to external script
- Improve code readability and reusability for future notification enhancements
- Replace printf and jq command chain with Python script for payload generation
- Improve readability by using Python string concatenation instead of nested printf format specifiers
- Ensure proper JSON encoding with ensure_ascii=False to preserve Chinese characters
- Simplify environment variable interpolation using os.environ dictionary access
- Add GitHub Actions workflow to notify WeChat on release branch merges
- Implement multi-step pipeline: sync ref, verify latest PR, fetch commits
- Integrate Aliyun Qwen AI for automated Chinese commit message summarization
- Send formatted Markdown notifications to WeChat webhook with release details
- Include branch, author, PR title, AI summary, and PR link in notifications
- Add redbear-mem-benchmark directory to .gitignore
- Prevents benchmark artifacts from being tracked in version control
- Aligns with existing pattern of ignoring redbear-mem-metrics directory
- Add GitHub Actions workflow to notify on merged release branch PRs
- Implement HEAD sync check to ensure branch is up-to-date before notification
- Fetch commit messages from merged PR for AI summarization
- Integrate Alibaba Qwen AI to generate Chinese release summaries for QA team
- Send formatted Markdown notifications to WeChat webhook with PR details and AI summary
- Workflow triggers only on final PR merge to release branches to avoid duplicate notifications
- Replace Statement-based implicit memory count (count/3) with actual
MemorySummary node count filtered by DERIVED_FROM_STATEMENT relationship
- Add minimum threshold of 5 MemorySummary nodes before reporting data
- Add _build_empty_profile() to return structured empty profile when
insufficient data exists, skipping unnecessary LLM calls
Added `get_user_by_id_regardless_active` in user repository to support activation/deactivation workflows, updated `user_service` to use it, and refactored `_enrich_release_config` in `app_dsl_service` to accept `default_model_config_id` as a parameter instead of reading from config dict.
