Claude tool landscape

Claude Code vs. Claude Chat vs. Claude Cowork — what's the difference and when do I use each?

By Mika Reyes and Nick Reyes ·

Short answer

Use Claude Chat for quick questions and drafts in the browser, Claude Cowork when you want Claude to read local files without a terminal, and Claude Code when you want an agent that works inside a project folder, edits files, runs skills, and connects to tools through MCPs.

The longer answer

Three surfaces, three habits: browser chat, desktop with local files, and agent-style work inside a project folder.

Claude Chat (claude.ai) is the browser experience you probably already know. Use it for drafting, summarizing uploads, Q&A, and rough brainstorming. Projects there keep instructions and shared context for chats inside that project.

Claude Cowork (desktop app) runs on your machine so Claude can read local files and see what you have open, without you uploading everything. Good for briefings from desktop docs and lighter automation when you do not want a terminal.

Claude Code is the heavier-duty option (terminal or desktop): it works inside one project directory, reads and writes files there, runs workflows, wires MCPs, and is where skills and repeatable agent setups tend to live.

Rough split:

  • Quick question or one-off → Chat
  • Local files, no terminal → desktop Cowork
  • Reusable workflows, skills, MCPs → Claude Code

Learn Claude Code hands-on

These answers come from our Maven Lightning Lesson Q&A. To build your own agent setup with CLAUDE.md, skills, and MCPs step by step, join Master Claude Code as a Non-Technical Pro or browse all Claude Code for Non-Technical Pros guides.

Using an AI assistant? Give it the Markdown version of this answer.