# Agentic AI Systems

A reference collection of runnable, checked Python implementations of agent design patterns such as prompt chaining, routing, parallelization and multi-window autonomous agents.

Agentic AI Systems documents agent orchestration patterns and pairs each one with a runnable code file that is checked in continuous integration, rather than only prose. It groups patterns into predefined workflows where code controls the flow, such as baseline, prompt chaining, routing, parallelization, orchestrator-worker and evaluator, and autonomous-agent patterns where the model controls the flow, plus implementation notes for building the pieces, such as subagents, commands, skills and hooks, in Claude Code.

A companion section tracks what changed in agent development between 2025 and 2026, covering context engineering, the multi-agent debate, and the AGENTS.md, SKILL.md and MCP stack. A selection guide and glossary help a reader pick the pattern that matches a given problem, and some patterns are said to run on a live schedule as a demonstration.

Website: https://github.com/ThibautMelen/agentic-ai-systems