Normalize OpenAI-style and LangChain standard image blocks into LangChain
standard content blocks so provider block_translators auto-convert for
either OpenAI or Anthropic. Flatten multimodal content to plain text when
persisting history and computing term embeddings.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The local file-parsing pipeline (upload -> Huey async parse -> generate
projects/data/.../document.txt) is no longer needed: RAG retrieval runs
against the backend vector store and does not read the local parse output,
so removing this has zero impact on existing bot Q&A.
- Delete task_queue/ (Huey queue, consumer, tasks, task status store)
- Delete parsing utils: dataset_manager, single_file_processor,
data_merger, project_manager
- Delete db_manager.py (only managed task_status.db)
- routes/files.py: keep only POST /api/v1/upload; drop all
parse/queue/task endpoints
- routes/projects.py: drop /tasks endpoint and task_status import
- utils/__init__.py & api_models.py: remove exports/models for deleted
modules and queue task models
- start_unified.py & start_all_optimized.sh: no longer launch the
queue consumer
- Drop huey dependency (keep redis)
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Remove routes/mcp_resources.py (ui:// URI resolver endpoint)
- Frontend now directly accesses /robots/{bot_id}/skills/{server}/apps/{resource}.html
- Add Daytona fallback middleware to fetch files from sandbox on 404
- Add utils/daytona_file_fetcher.py for on-demand single file download
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Update langfuse CallbackHandler to use trace_context instead of removed trace_id/session_id/user_id params
- Pass session_id/user_id via LangChain metadata with langfuse_ prefix
- Move dereference param from TarFile.add() to tarfile.open()
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New version of NewAPI masks token keys with asterisks in list/detail endpoints.
Added get_token_key() method that calls the dedicated endpoint to retrieve
the full key, and integrated it into get_or_create_token() flow.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The incremental sync used `find -type f` which misses symlinks (type l),
so dataset symlinks were never detected and synced to the sandbox.
Additionally, `tar.add()` without `dereference=True` would store broken
symlinks pointing to host-only paths.
- _list_local_changed_files: match both regular files and symlinks
- _tar_workspace_entries: dereference symlinks to pack actual content
- Unify dataset path to `datasets/` (plural) in prompts and SKILL.md
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Convert all Chinese comments, docstrings, logger/print output,
HTTPException detail messages, and API response messages to English
across the entire codebase. Functional zh/ja localized strings
(e.g. prompt templates, timezone display names, date formats) are
preserved as-is.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
