Skip to content

Agent Guides

Comprehensive, production-ready documentation for building AI agents across 16+ frameworks.

Welcome to the Agent Guides collection—your complete resource for mastering AI agent development. Whether you’re building your first chatbot or deploying enterprise-scale multi-agent systems, you’ll find detailed guides, practical recipes, and production patterns here.


Multi-agent systems with lightweight primitives and powerful abstractions.

Quick Links:

Install: pip install openai-agents


Lightweight Python framework where agents write code instead of JSON.

Quick Links:

Install: pip install smolagents


  • Chatbots & Assistants: OpenAI Agents SDK, SmolAgents, CrewAI
  • Data & Retrieval: LlamaIndex, Haystack
  • Complex Workflows: LangGraph, AG2
  • Cloud Native: Amazon Bedrock, Microsoft Agent Framework
  • Type Safety: PydanticAI
  • Python: All frameworks
  • TypeScript: LangGraph, Claude SDK, OpenAI SDK
  • C#: Semantic Kernel
  • Beginner: SmolAgents, PydanticAI, CrewAI
  • Intermediate: LangGraph, LlamaIndex, OpenAI Agents SDK
  • Advanced: AG2, Haystack, Custom multi-agent systems

View All Guides →


Every framework guide contains:

  1. README - Quick start & overview
  2. Comprehensive Guide - Complete feature reference (~50-100 pages)
  3. Production Guide - Deployment & scaling
  4. Diagrams Guide - Architecture visualisations
  5. Recipes - Copy-paste ready code examples

Use the table below or browse above:

FrameworkBest ForComplexityTime to Learn
SmolAgentsSimple agentsLow1 hour
PydanticAIType safetyLow-Medium2 hours
OpenAI Agents SDKMulti-agentMedium4 hours
CrewAIRole-based teamsMedium4 hours
LangGraphComplex workflowsHigh8 hours
LlamaIndexRAG systemsMedium6 hours

Each guide folder contains a README with:

  • Installation instructions
  • Basic example
  • Key concepts
  • Links to deeper documentation

The Recipes guide contains copy-paste ready examples:

  • Common use cases
  • Integration patterns
  • Best practices in action

For production systems, read:

  • Production Guide
  • Diagrams Guide for architecture
  • Security & monitoring sections

Autonomous systems with instructions, tools, and decision-making.

Executable actions agents can call to interact with external systems.

Coordination of multiple specialised agents:

  • Sequential handoffs
  • Parallel execution
  • Hierarchical delegation

Safety constraints:

  • Input validation
  • Output filtering
  • Rate limiting
  • Access control

State management:

  • Conversation history
  • Long-term storage
  • Context management

Want to compare features across frameworks? Check out our Comparison Guide.


  1. Check the framework’s README
  2. Search the Comprehensive Guide
  3. Look for examples in Recipes
  4. Review the Production Guide for deployment issues
  • Report documentation errors
  • Suggest improvements
  • Contribute examples

Help make this resource better:

  • Fix errors or typos
  • Add new examples
  • Improve diagrams
  • Update for new versions
  • Share your patterns

Contribution Guidelines →


All guides follow consistent standards:

  • British English spelling (optimisation, favour, analyse)
  • Tested code examples
  • Production-ready patterns
  • Clear structure from basic to advanced
  • Visual diagrams for architecture

April 2026


This collection is maintained by the community for the community. Thank you to all contributors!


Start exploring: Browse All Guides →