- Updated the CSS classes for the AI assistant bar to adjust positioning based on the new side panel width variable.
- Introduced a new `onWidthPreview` prop in the SidePanelRail component to handle live resizing and update the side panel width dynamically.
- Added width state management in the SidePanelStore to ensure proper width handling during panel interactions.
- Improved layout handling in the course layout component to accommodate the new side panel width adjustments.
- Implemented a new endpoint to delete failed manual p1080 renditions, restoring the master manifest and updating metadata.
- Enhanced the HLS upload process with better error handling and cleanup for aborted uploads.
- Introduced tests for HLS encoder functionality to ensure correct selection of rungs based on video source properties.
- Updated video upload component to manage reserved asset IDs and perform cleanup on abo
Add get_lesson_transcript (teacher) and read_lesson_transcript (student)
tools that fetch a lesson's uploaded-video transcript from media_transcript
by the video assetIds. The agents call them only when a question is about
the video's spoken content, so text-only chats keep a cacheable prefix.
- Introduced a concurrency gate to limit parallel uploads, preventing R2 overload.
- Implemented batching for presign requests, allowing up to 1500 paths per request.
- Added content type mapping for HLS uploads to ensure correct MIME types.
- Refactored upload logic to utilize the new concurrency and presign features, improving reliability and performance.
- Replaced email/token-based authentication with HTTP Basic Auth for the queue dashboard in production.
- Updated course components to use courseId instead of courseSlug for better consistency.
- Adjusted role-based security checks to utilize isStudentExperience for conditional rendering.
- Enhanced view-as-student functionality to redirect correctly based on courseId.
These were used by the redis client and document-parsing services that moved
from apps/api into packages/core, but the dependencies stayed in apps/api,
breaking the @cio/core build with TS2307 module-not-found errors. Move them
(and @types/pdf-parse) to where the code now lives.
New features:
- HLS video encoding/streaming (hls routes, hls-encoder) and a media
transcripts pipeline (transcripts route + media-transcript queries)
- AI assistant chat vs. run model: agent-course-generation worker/queue,
run-monitor pane, chat-changes card, content ask-ai bar
- Per-theme course-page previews across org landing-page themes (bold,
classic, corporate, editorial, etc.) with a shared course token/component set
- View-as-student mode and login-link / link-invite invite flows
Refactor:
- Extract packages/core consolidating config (env, storage, dashboard-url),
services and utils previously duplicated under apps/api/src/config
- Remove legacy agent service modules (chat-tools, chat-context,
context-window, document, fetch-url, lesson-content)
- Relax Cross-Origin-Resource-Policy to allow cross-origin media embedding
- Add format script to tenant-router so turbo run format covers it
Resolve sessions before rate limiting so keys are per-user, exempt trusted server API key calls and get-session, and raise the GET /organization budget for layout loads.
Co-authored-by: Cursor <cursoragent@cursor.com>
Rotimi Best