Where AI Agent Concepts Come to Life
An interactive learning platform for mastering AI agents, patterns, and production architectures

** Live Platform:** openagentschool.org

Open Agent School is a comprehensive educational platform designed to help developers, architects, and product managers master AI agentsfrom foundational concepts to advanced production patterns.

Core Foundations

• AI Agent Fundamentals - Architecture, reasoning, tool use, and memory systems
• Agent Communication Protocols - MCP (Model Context Protocol) and ACP (Agent Communication Protocol)
• AI Safety & Governance - Responsible AI practices, security, and compliance frameworks
• Evaluation Methods - Testing, benchmarking, and quality assurance for agents

Advanced Patterns

• Multi-Agent Systems - Coordination, orchestration, and swarm intelligence
• Data Autonomy Patterns - Perception normalization, schema decomposition, action grounding
• Business Use Cases - Commerce, robotics, healthcare, and enterprise applications
• Educational Frameworks - Socratic learning, adaptive tutoring, and skill assessment

Production Practices

• Deployment & Operations - Platform ops, monitoring, and continuous improvement
• Cost & Value Engineering - Optimization strategies and ROI frameworks
• Human-Centric Adoption - Change management, HITL workflows, and recovery design
• Ecosystem & Partnerships - Vendor evaluation and integration stewardship

• 52+ Agent Patterns with code examples and visualizations
• Learning Atlas - Interactive taxonomy tree with exportable diagrams
• Study Mode - Socratic questioning and discovery-based learning
• Critical Thinking Challenges - Real-world scenarios with AI-powered assessment
• Knowledge Quiz - Test comprehension across all topics

• D3 Visualizations - Dynamic, interactive diagrams for complex concepts
• Algorithm Walkthroughs - Step-by-step execution visualizations
• Export Suite - SVG, PNG (2x/3x), and PDF export for presentations
• Dual-Mode Views - Simple and technical modes for different learning levels

• Code Playbooks - Runnable code examples in Python and TypeScript
• Interactive Scenarios - Debug challenges and failure recovery exercises
• Live Pattern Runners - Execute patterns in-browser with real output
• Strategy Toolkits - Downloadable canvases and frameworks (XLSX format)

• Offline Access - Content available without internet
• Install App - Native app experience on mobile and desktop (menu: Install App)
• Responsive Design - Optimized for all screen sizes
• Dark/Light Themes - Comfortable reading in any environment

1. Visit: openagentschool.org
2. Explore: Browse Core Concepts or Agent Patterns
3. Learn: Use Study Mode for guided discovery
4. Practice: Try interactive scenarios and quizzes

`�ash

git clone https://github.com/bhakthan/OpenAgentSchool.git cd OpenAgentSchool

npm install

npm run dev

`

`powershell

.\start-backend.ps1

`

Note: The frontend works standalone. Backend services enable advanced features like user progress tracking and AI-powered assessments.

• Detailed Documentation - Comprehensive feature list and architecture
• Backend Integration - Microservices setup and API reference
• Agent Guide - Guide for AI coding agents contributing to the project
• Architecture Overview - System design and service dependencies

PWA Install Experience

• Disabled auto-popup installation prompt (was appearing after 3-5 seconds)
• Added "Install App" menu item in top navigation
• Platform-specific installation dialogs (iOS vs Android/Desktop)
• One-click install for supported browsers

Console Cleanup

• Disabled backend health checks to prevent CORS errors when services aren't running
• Suppressed Workbox debug logs in development mode
• Fixed React ref forwarding warnings in FutureStateTrends component

PWA Manifest Enhancement

• Added all 8 icon sizes (72x72 through 512x512) to eliminate browser warnings
• Proper support for different device screen densities
• Consistent manifest across dev and production

Code Quality

• Fixed component ref handling to follow React best practices
• Improved error handling in OfflineBanner component
• Enhanced Vite configuration for cleaner development logs

Documentation Restructure

• Moved comprehensive README to docs/DETAILED_README.md
• Created simplified main README for better first impression
• Preserved all technical details in dedicated documentation

The platform's top navigation follows a deliberate learning progression:

1. Core Concepts → Foundation first (agents, prompting, evaluation, security)
2. Agent Patterns → Architectural patterns and best practices
3. Applied AI Skills → Hands-on skills application
4. Adoption Playbook → Organizational transformation strategies (after learning the concepts)
5. Azure Services → Platform-specific implementation tooling
6. Learning Atlas → Visual reference and taxonomy
7. Study Mode → Practice and Socratic discovery

Rationale: Learners progress from Learn (concepts) → Build (patterns) → Apply (skills) → Adopt (organizational strategy) → Implement (platform) → Practice (study/quiz). The Adoption Playbook is positioned after foundational learning so that strategic thinking is grounded in technical understanding.

1. Core Concepts Agent Fundamentals
2. Learning Journey Map (interactive progression)
3. Knowledge Quiz (assess comprehension)
4. Study Mode: Socratic Questions

1. Agent Patterns Educational Patterns
2. Multi-Agent Systems
3. Data Strategy Visualization
4. Code Playbooks & Examples

1. Data Autonomy Patterns (8 foundational patterns)
2. Critical Thinking Challenges
3. Production Operations & Monitoring
4. Strategy Toolkits (downloadable canvases)

1. Adoption Playbook → Mission Brief (discovery questions)
2. Transformation Journey (5-phase roadmap)
3. Operating System Design (governance, platform, culture)
4. Activation Plan (first 90 days)
5. Study Mode: Adoption Strategy challenges

We welcome contributions! Whether it's:

• Bug Reports - Found an issue? Open an issue
• Feature Requests - Have an idea? We'd love to hear it!
• Documentation - Help improve our guides
• Code Contributions - Submit a pull request

See AGENTS.md for guidance on contributing code.

• 35+ Agent Patterns (Educational + Data Autonomy + Business Use Cases)
• 150+ Quiz Questions across 12 categories
• 40+ Interactive Visualizations (D3, React Flow, custom diagrams)
• 8 Strategy Toolkits (downloadable XLSX canvases)
• 4 Learning Modes (Concepts, Study, Critical Thinking, Interactive Scenarios)

Frontend

• React 18 + TypeScript + Vite
• TailwindCSS 4 + Radix UI
• D3.js for visualizations
• React Flow for agent workflows

Backend (Optional)

• Python 3.12 + FastAPI
• PostgreSQL + DuckDB
• OpenAI/Anthropic/Azure OpenAI
• Docker Compose

Infrastructure

• Azure Static Web Apps
• GitHub Actions CI/CD
• Progressive Web App (PWA)
• Service Worker caching

MIT License - see LICENSE for details.

Built with by the Open Agent School community, powered by AI assistance and lots of human expertise.

Special Thanks:

• Microsoft Learn AI documentation
• OpenAI, Anthropic, Google, and Azure AI teams
• Open source community (React, D3, Vite, and countless libraries)

• Website: openagentschool.org
• GitHub: github.com/bhakthan/OpenAgentSchool
• Issues: Report bugs or request features