Explore, plan, code, commit on existing code
Asking an agent to change code it doesn't know often yields a result that looks plausible but misses the point. This lesson teaches you to run a real task with Claude Code in a structured way, from understanding the code to the commit, checking the plan before any line is written and keeping a way to roll back.
Lesson objective
By the end of this lesson, you will be able to run a non-trivial task on existing code in four steps (explore, plan, code, commit), with plan mode, verification criteria the agent runs and Git commits, and to know when checkpoints are enough and when they are not.
Topics covered
- Claude Code plan mode
- AI coding agent
- existing codebase
- Git commits
- AI-assisted workflow
Where it fits
Plan, then delegate with method
How do I run a real task without the agent heading the wrong way, and which tool do I give it for what?
Lessons in this module
- Explore, plan, code, commit on existing code (this lesson)
- Subagents, skills, commands and MCP servers: the right tool for each need
What you will learn in this module
This lesson is part of the course Work with Claude Code on a real project: context, planning, review
- Run a task on existing code as explore, plan, code, commit, with plan mode, verification criteria and Git rather than checkpoints alone.
- Choose and justify the right mechanism for a need (CLAUDE.md, skill or command, subagent, MCP server, hook) and set it up in the repository.
Related courses
- AI & DevelopmentJuniorSome basics are useful: introductory knowledge of the topic.
Build and ship a web app with an AI coding agent
2 hr 45 min
- AI & DevelopmentAdvancedRegular practice: you already use the tools or work on this topic.
Add a database, authentication and payments to your app
4 hr
- Product ManagementAdvancedRegular practice: you already use the tools or work on this topic.
Git for PMs: ship as a team without putting production at risk
3 hr 30 min