Guide
Theme
Agentic Coding
This theme covers the practical craft of working with AI coding agents: structuring prompts for real codebases, choosing between agent tools, reviewing and shipping AI-written code, and using MCP and tool-use patterns to extend what an agent can do.
- Working with Claude Code, Cursor, Aider, and similar agents
- Agent-friendly repo conventions and reviewing AI-written code
- MCP and tool-use patterns for real codebases
Start here
New to this theme? Read these in order
Each one sets up the next.
01
Vibe-coding with AI: a pragmatic first project
Vibe-coding is the loose, conversational style of building software with an AI pair programmer doing the keystroke-level work while you stay focused on intent. It's wonderful for shipping a first project, and it has a few sharp edges that are worth understanding before you start.
Read this guide02
Structuring an agent prompt for a real codebase
Most failed agentic coding sessions come from the same place: the prompt was vague, the agent inferred the wrong thing, and you spent thirty minutes chasing AI-generated bugs that wouldn't exist if the first message had been precise. A short structure for the opening prompt: applied consistently; turns the agent from an unpredictable junior into a useful pair partner.
Read this guide03
Planning a task the agent can actually finish
Most coding agents fail on vague tasks, not hard code. Learn a four-part task spec: scope, constraints, verification, size, that lets agents ship mergeable changes.
Read this guide04
Reviewing AI-written code: a checklist for what to scrutinise
Code generated by an AI agent looks like code, runs like code, and passes the obvious tests like code, but it has its own characteristic failure modes. Reviewing AI output well isn't about distrusting it; it's about knowing which categories of mistake are most likely and looking for them deliberately. The review skill compounds quickly once you know where to focus.
Read this guide05
Agent-friendly repo conventions: AGENTS.md, structure, tests
A codebase that's easy for agents to work in is also a codebase that's easy for new humans to work in. Most of the conventions that pay off, clear structure, fast tests, named patterns, a short conventions document, were good ideas before AI coding. Agent-assisted work just makes the cost of skipping them visible faster.
Read this guideGuides
Agentic Coding: guides worth reading
Start with one of these, then use the broader theme to decide what to read or try next.
Guide
Planning a task the agent can actually finish
Read guideGuide
Agent-friendly repo conventions: AGENTS.md, structure, tests
Read guideGuide
Reviewing AI-written code: a checklist for what to scrutinise
Read guideGuide
Structuring an agent prompt for a real codebase
Read guideGuide
Vibe-coding with AI: a pragmatic first project
Read guideTaim.io app
Keep going with agentic coding in the app
Set this as your objective and the coach turns what you have read into regular, hands-on practice.
What would you like to get really good at?