Skip to content

drdgreed/lesson-planning-assistant

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

Lesson Planning Assistant - Prototype

An AI-powered lesson planning tool for K-12 teachers using Claude 3.5 Sonnet.

Overview

  • Backend: FastAPI + SQLite + Claude API
  • Frontend: React + Axios
  • Demo Accounts: 3 teachers (Grade 4 Math, Grade 6 ELA, Grade 9 Science)
  • Generation Time: 30-60 seconds per lesson
  • Expected Quality: 75% usable foundation requiring teacher refinement

Features

✅ User authentication (simple token-based) ✅ Standards-aligned lesson generation (14 sample standards) ✅ 5 lesson templates (Introducing New Concept, Practice & Review, etc.) ✅ Differentiation scaffolds (struggling learners, advanced, ELL) ✅ Save lessons to database ✅ Export lessons as text files ✅ 4-Pass Refinement Workflow guidance

Setup Instructions

Prerequisites

  • Python 3.8+
  • Node.js 14+
  • Anthropic API key

Backend Setup

  1. Navigate to backend folder:
   cd backend
  1. Create virtual environment:
   python -m venv venv
   source venv/bin/activate  # On Mac/Linux
   # venv\Scripts\activate  # On Windows
  1. Install dependencies:
   pip install -r requirements.txt
  1. Initialize database:
   python init_db.py
  1. Set API key:
   export ANTHROPIC_API_KEY="your-api-key-here"
  1. Start backend server:
   python main.py

Server runs on http://localhost:8000

Frontend Setup

  1. Navigate to frontend folder:
   cd frontend
  1. Install dependencies:
   npm install
  1. Start frontend:
   npm start

App opens at http://localhost:3000

Demo Accounts

Username Password Teacher Name Grade Subject
teacher1 password1 Sarah Johnson 4 Math
teacher2 password2 Michael Chen 6 ELA
teacher3 password3 Dr. Emily Rodriguez 9 Science

Usage Flow

  1. Login with a demo account
  2. Fill lesson context form:
    • Grade level and subject
    • Learning standard (auto-filtered)
    • Learning objective
    • Duration, student context, resources
    • Optional: Select lesson template
  3. Generate lesson (wait 30-60 seconds)
  4. Review generated plan with 4-Pass refinement guidance
  5. Save or export the lesson

API Endpoints

  • POST /api/login - Authenticate user
  • GET /api/standards - Get learning standards (filterable)
  • POST /api/generate_lesson - Generate lesson with Claude
  • POST /api/save_lesson - Save lesson to database
  • GET /api/my_lessons - Get user's saved lessons
  • GET /api/export_docx/{lesson_id} - Export lesson as DOCX

Project Structure

lesson-planning-assistant/
├── backend/
│   ├── main.py              # FastAPI application
│   ├── init_db.py           # Database initialization
│   ├── standards_db.json    # 14 sample standards
│   ├── database.db          # SQLite database
│   └── requirements.txt     # Python dependencies
├── frontend/
│   ├── src/
│   │   ├── components/      # React components
│   │   ├── utils/           # API client
│   │   ├── App.js           # Main app component
│   │   └── App.css          # Global styles
│   └── package.json         # Node dependencies
└── README.md

Tech Stack

Backend:

  • FastAPI - Modern Python web framework
  • SQLite - Lightweight database
  • Anthropic Claude API - AI lesson generation
  • python-docx - Document export

Frontend:

  • React - UI library
  • Axios - HTTP client
  • CSS3 - Styling

Limitations (Prototype Scope)

  • Localhost only (no deployment)
  • Simple authentication (not production-ready)
  • 14 hardcoded standards (not full standards database)
  • Text export only (DOCX export simplified)
  • No version history or collaboration features
  • No mobile optimization

Success Criteria

✅ All demo accounts can log in ✅ Lesson generation completes in <90 seconds ✅ All lesson sections display correctly ✅ Save and export functions work ✅ 5-7 minute demo runs smoothly

Realistic Expectations

What This Proves:

  • AI can generate structured lesson drafts quickly
  • Teachers can refine in reasonable time (45-70 min)
  • Net time savings: 1.5-2 hours per lesson

What This Does NOT Prove:

  • 95% classroom-ready quality (target: 75% usable foundation)
  • High adoption rates (realistic target: 35-45%)
  • Works for all teachers (best for 2-5 years experience)

Next Steps (Phase 2)

  • User testing with 10-15 real teachers
  • Full standards database integration
  • Enhanced DOCX export with proper formatting
  • Save/load functionality improvements
  • Mobile responsive design
  • Production deployment
  • Analytics and usage tracking

License

Prototype for evaluation purposes.

Contact

For questions or feedback about this prototype, please contact the development team.

About

AI-powered K-12 lesson planning tool using Claude API

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors