Claude Code + Cowork

What is the difference between Claude Code and Claude Cowork?

Claude Code is an agentic coding tool you drive from the terminal, IDE, or web — it reasons about your whole project, plans changes, and writes and runs code. Claude Cowork uses the same engine but is built for teams: shared project context, background agents, and standards applied across everyone's work. If you know Code, you already know most of Cowork.

What is auto-mode in Claude Code?

Auto-mode lets Claude run autonomously — reading, writing, testing, and iterating without stopping to ask permission at every step. It is ideal for framework migrations, boilerplate generation, test coverage, and bulk refactors. You still review the output before merging.

Can Claude run tasks on a schedule?

Yes. With scheduled tasks you describe a job once, pick a cadence such as daily, weekly, or hourly, and Claude runs it automatically with access to your connectors. Today it runs on your own machine, so your computer needs to be awake with the app open for scheduled runs to fire.

What is MCP and why does it matter?

MCP (Model Context Protocol) is the open standard that lets Claude connect to your real tools and data — databases, internal APIs, Notion, Slack, GitHub, and more. Sharing MCP configuration across the team means everyone gets the same integrations without per-person setup.

Do you need to be a developer to use Cowork?

No. Although Cowork is built on a coding tool, it is designed so any team can automate everyday work — reports, digests, triage, and research — by describing tasks in plain English.