What this is

This is the agent-shaped cut of the AI Path: every article about building, extending, scaling, and operating agents, sorted into the order the ideas actually depend on each other. The broader AI stack map covers everything around it - the engine, the models and platforms, protocols and retrieval, and the shipped tools; this page goes deep on the one layer where teams now spend most of their engineering time, and lists only the articles that belong to it.

The spine of the map is a single idea: an agent is a model in a loop, and everything about that loop - what it can do, what it costs, how it fails, when to split it into several - follows from the one block of text it sends the model every turn. Read the layers in order and each article has what it needs from the one before.

The agents are the easy part. The window, the boundaries, and the budgets are the system.- the theme this map keeps returning to

The idea

What an agent is and what "agentic" means. Three articles - the definition, the system view, and the loop stage by stage - with the engine underneath covered in the AI stack.

LAYER 01

The idea

AI · Agents

Start with the definition - model, tools, loop - then the concept piece, which steps back to autonomy and where agents earn their keep. The loop article takes the control loop apart stage by stage and is the bridge to every layer below. If models themselves are new to you, read What is an LLM? in the AI stack first: an agent inherits every property of the engine it wraps, statelessness above all.

The window

Two articles, and the most load-bearing layer on this map. The first is the window itself; the second separates the four words - working memory, state, long-term memory, retrieval - that builders collapse into "memory". Every layer after it is secretly about the same thing.

LAYER 02

The window

AI · Context

Every turn, an agent sends the model one block of text and gets one back. What goes into that block, who puts it there, what each part costs, and what gets thrown away when it stops fitting - this is the mechanical foundation for everything below: extensions manage the window, architectures split it, production budgets pay for it.

Extending an agent

Functions, MCP servers, Skills, hooks, plugins - five words for attaching capability to an agent, each with a different relationship to the window.

LAYER 03

Extensions

AI · Extensions

Start with the three-way comparison to get the vocabularies straight, then the packaging formats. The protocol that carries tools, MCP, has its own layer in the AI stack; here it appears as one mechanism among four. The four-mechanism piece is the one to send a teammate who keeps mixing them up; the plugins article covers how Claude Code distributes all of it as one install.

Architectures

The step past one agent: whether to take it, and the coordination patterns waiting on the other side.

LAYER 04

Architectures

AI · Orchestration

Read them in this order. The systems piece is the decision: why one agent stops scaling, what handoffs cost, and when a single agent is still right. The patterns piece is the catalog for after the decision: orchestrator-workers, pipelines, fan-out, critique, and routing, with a rule for choosing.

Running in production

Building an agent is a week. Operating one is the job. Three articles cover the operating side, and they are best read before the first deployment.

LAYER 05

Production

AI · Operations

Evaluation is how you know the agent works before trusting it; the failure piece names the recurring ways long runs go wrong; the cost piece explains why a run bills many times what its prompt suggests once the whole context is re-sent every turn.

Frameworks

The harnesses you would build an agent with. The two shipped agents worth studying as reference implementations - Claude Code and Claude Cowork - are products, and live in the AI stack's tools layers.

LAYER 06

Frameworks

AI · Frameworks

The survey covers the libraries and platforms a builder is likely to pick; the Microsoft Agent Framework piece looks closely at one mainstream option. For the same architecture as a shipped product, read Claude Code and Claude Cowork in the AI stack.

Three reading orders

The map is sorted by layer, but most readers want a route. Three orders, depending on where you stand.

The map at a glance

If you read only this section, here is the agentic stack from the idea up:

Every article in the map

Every article on the agentic slice, by layer, with a one-line reason to read.

Layer Article Why read it
Idea What is an agent? 101 The strict definition - model, tools, loop - a chatbot, a workflow, and an agent told apart, and one task traced end to end.
Idea What is agentic AI? 101 The concept - what makes a system "agentic," the parts of an agent, and how it differs from a chatbot.
Idea Inside an AI agent: the agent loop 201 The control loop stage by stage - state, planning, tool selection, observation, recovery, termination, approval - and why an agent is not an LLM plus tools.
Window What's in an agent's context? 101 The one block of text an agent sends every turn - what is in it, what it costs, what falls out first.
Window Inside agent memory 201 Working memory, state, long-term memory, and retrieval separated - the four things "add memory" actually means.
Extensions Functions, MCP, and Skills 101 The three vocabularies for "things an agent can do," compared and untangled.
Extensions What are Agent Skills? 201 Anthropic's packaging format for instructions an agent loads only when the job calls for them.
Extensions Skills, MCP, Hooks, and Plugins 201 The four extension mechanisms side by side, and which job each one belongs to.
Extensions What are Claude Code plugins? 201 Claude Code's packaging for sharing skills, commands, and MCP servers as one install.
Architecture Multi-agent systems 201 When to split one agent into several, the three shapes that work, and what handoffs actually cost.
Architecture Multi-agent orchestration patterns 201 Orchestrator-workers, pipelines, fan-out, critique, routing - and how to pick the one your job needs.
Production How to evaluate an agent 301 How to measure an agent before trusting it with real work.
Production Why agents fail 301 The recurring failure modes of long agent runs, and the guards that catch them early.
Production What an agent actually costs 201 Why a run costs many times what its prompt suggests once the whole context is re-sent every turn.
Frameworks AI agent frameworks 101 A survey of the libraries and platforms used to build agents in production.
Frameworks What is Microsoft Agent Framework? 201 Closer look at one mainstream option, with the surrounding Microsoft ecosystem context.