尚未收集儲存庫資訊。
尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
依最近已完成的 8 小時觀測回合間的 GitHub 星號增量排序
The learning platform for companies, colleges and training institutes. Free and open source, with an Enterprise edition.
翻譯準備中
ESP32-S3 N16R8 bench guide: live pinout, lessons, simulators and the bench log.
翻譯準備中
關於自動駕駛車輛控制演算法的 Python 示例程式碼與文件。本專案可作為技術指南書，供初學者學習這些演算法與軟體架構。
AI 翻譯
A study guide/notes in a click of a button. Scrapes data from page including frontend text and slideshows and creates notes/study guide from it. Includes AI chatbot to answer any questions from scraped data. Also includes "example" option to provide example of specified concept in AI chatbot.
翻譯準備中
取自儲存庫 README
正在載入完整 README。
檢視完整 README ↗A Chrome side-panel extension and Classroom web app that share one Cordia Tutor session for capturing and learning from student-approved educational content.
CordiaClassroom captures content from learning platforms, online textbooks, PDFs, slideshows, and webpages, then turns it into organized study materials.
| Feature | Description | AI Usage |
|---|---|---|
| Smart Notes | AI extracts key information, filtering out navigation and irrelevant content | GPT-4o identifies important concepts, definitions, and facts |
| Study Guide | Generates comprehensive Q&A study guides | GPT-4o creates varied question types (definition, explanation, application, comparison) with detailed answers |
| Flashcards | Creates flashcard-style question/answer pairs for quick review | GPT-4o generates concise Q&A pairs optimized for memorization |
| Chat Assistant | Interactive Q&A using captured content as knowledge base | GPT-4o answers questions in 3 modes: short, detailed, or with examples |
CordiaClassroom automatically detects and extracts content from slideshows:
| Platform | Detection | Extraction |
|---|---|---|
| Canvas LMS | Native slideshow modules | Full text extraction from all slides |
| Google Slides | Embedded iframes | Detects embed, prompts to open directly |
| PowerPoint Online | OneDrive/SharePoint embeds | Detects embed, extracts surrounding content |
| PowerPoint Files | .pptx download links | Downloads and parses slide content |
| Prezi | Embedded presentations | Detects embed URL |
| Generic Slideshows | CSS class detection (.slides, .carousel) | Extracts visible slide content |
┌─────────────────────────────────────────────────────────────────┐
│ Chrome Extension │
├─────────────────────────────────────────────────────────────────┤
│ Side panel UI │ content.js │ background.js │
│ - Side-panel UI │ - Slideshow detect │ - API calls │
│ - Display results │ - PDF extraction │ - Message router │
│ - Chat interface │ - LMS content grab │ │
│ - Flashcard view │ - PPTX detection │ │
└─────────────────────────────────────────────────────────────────┘
│
▼ Authenticated HTTP
┌─────────────────────────────────────────────────────────────────┐
│ FastAPI Backend │
├─────────────────────────────────────────────────────────────────┤
│ /ingest │ /generate │
│ - Review sections │ - AI notes and study guides │
│ - Detect slideshow │ - AI flashcards │
│ - Return metadata │ │
├─────────────────────────────────────────────────────────────────┤
│ /chat │ Text Processing │ Tutor session │
│ - Shared skills │ - LMS filtering │ - One writer │
│ - Context-aware │ - Smart chunking │ - Action evidence│
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ OpenAI GPT-4o │
│ - Intelligent note extraction (identifies exam-worthy content) │
│ - Study guide generation (varied question types) │
│ - Flashcard creation (memorization-optimized) │
│ - Contextual Q&A (short/detailed/example modes) │
│ - Slideshow summarization │
└─────────────────────────────────────────────────────────────────┘
| Feature | AI Role | Why AI? |
|---|---|---|
| Notes | Identifies key concepts, definitions, and exam-worthy facts | Human-like understanding of what's important to study |
| Study Guide | Generates varied question types with detailed answers | Creates questions that test understanding, not just memorization |
| Flashcards | Creates concise Q&A pairs for quick review | Optimizes content for memorization and recall |
| Chat | Answers questions using captured content as context | Provides personalized explanations based on the material |
| Slideshow Summary | Synthesizes slide content into coherent notes | Connects information across slides into study-ready format |
The content filter in services/text_processing.py automatically removes:
Supports: Canvas, Blackboard, Moodle, D2L Brightspace
In services/llm.py:
| Parameter | Value | Purpose |
|---|---|---|
model | gpt-4o | Best quality for educational content |
temperature | 0.3-0.5 | Balanced accuracy with some variety |
max_tokens | 150-2048 | Varies by feature (notes vs study guide) |
MIT