Mathews-Tom/Agentic-Design-Patterns resource

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1834 stars · 365 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A community-compiled repository containing the full text of the book 'Agentic Design Patterns' by Antonio Gulli and Mauro Sauco, organized as markdown chapters. It covers 21 core patterns for building autonomous AI agents, including prompt chaining, routing, planning, memory management, RAG, multi-agent orchestration, and MCP, with working code examples.

Use cases

• learn design patterns for building AI agents
• understand multi-agent orchestration techniques
• study prompt chaining and routing for LLM applications
• learn how to implement memory management in agents
• get practical examples of RAG and MCP integration
• prepare for building production-grade agentic systems
• find a reference on agent safety and error handling patterns

When to choose

• you are an engineer, researcher, or product manager moving beyond basic LLM API calls to build robust AI agents
• you want a free, comprehensive written reference with real code examples for agentic patterns
• you need coverage of production concerns like safety guardrails, evaluation, and human-in-the-loop design
• you want a structured curriculum of 21 patterns from foundational to multi-agent architectures

When to avoid

• you need runnable software or a framework rather than a book to read
• you are looking for a beginner introduction to calling LLM APIs
• you require a formally maintained codebase with releases and issue tracking, since this is compiled book content

Facets

learning-resource · maturity active

agent-framework prompt-engineering rag mcp large-language-models tutorials developer-tools book design-patterns multi-agent-systems llm-agents agentic-ai educational-resource ai-agents

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Mathews-Tom/Agentic-Design-Patterns")