Introduction
This series teaches how to build a coding agent using LangGraph. I called it Axon, a terminal-native coding agent like Claude Code, Manus, Codex, etc.
The aim is to make end to end coding agent design: state, nodes, edges, tool-calling mechanism, the ReAct loop, planning, sub-agent orchestration, memory, self-correction. Each chapter implements one of these as a working feature.
What we build: Axon#
Axon is a CLI coding agent. By the end of the series it will:
- Use tools such as: read, write, edit, and search files in any repo.
- Ask for confirmation before destructive changes.
- Plan multi-step tasks and execute them.
- Spawn sub-agents that work in parallel.
- Persist project-scoped memory across sessions.
- Self-correct on failed tool calls.
- Run on any hosted or local model.
Each chapter ships a runnable feature on top of a real codebase, not a notebook toy. The code is public, and every chapter links to the files it builds.
kubilaygulacdi/axonAxon, the coding agent this series builds. Each chapter links to the files it touches.This is the machine we end up with. Click around; you will have built every box yourself:
Why build it from scratch?#
It is about control over the state. LangGraph ships create_deep_agent and create_agent helpers that give you a standard agent in a few lines, but the moment you need your own logic inside a node, you are working around them instead of with them.
LangGraph is a low-level orchestration framework and runtime for building, managing, and deploying long-running, stateful agents so it allows us to maximum control for our domain specific agent.
The stack#
- LangGraph: state machine, everything we need to build AI agents
- LM Studio: local LLMs to serve
Chapter map#
| # | Chapter | What Axon gains |
|---|---|---|
| 0 | Introduction (this post) | Series scope, rules, stack |
| 1 | State and Graph | Minimal REPL, the agent skeleton |
| 2 | Tools and the ReAct Loop | Filesystem read, write, search |
| 3 | Human-in-the-Loop | Approval question for destructive operations |
| 4 | Planning and TODOs | A todo list the agent writes for itself |
| 5 | Subagents | Parallel research on a fresh context window |
| 6 | Memory | Long-term project-scoped memory across sessions |
| 7 | Self-correction | Reflexion + critic loop |
| 8 | Evaluation | Trajectory eval harness, cost dashboard |
Next#
We start with the smallest LangGraph that compiles: one node, one edge, a typed config, a working REPL. The four primitives introduced there - State, Reducer, Node, Edge - are what the rest of the series extends.