An AI workspace is one place where you run AI models or agents with the context and tools they need, so the work does not live in scattered chats. For coding agents, the free, MIT-licensed Friday desktop app is one: it runs your own agents side by side on your own computer.
What are the main types of AI workspace?
| Type | Built for | Typical work |
|---|---|---|
| Docs and wiki workspaces | Notes, projects and knowledge | Writing, summarizing, searching your docs |
| Multi-model chat workspaces | Using several AI models in one place | Questions, drafts, research |
| Agent and automation builders | Building agents that run a workflow | Support, sales and operations tasks |
| Coding agent workspaces | Running coding agents on a repository | Coding, review, research, reports |
Many people use more than one. The type matters because it decides what the workspace is good at.
What should an AI workspace have?
- Context the agent keeps, so you do not explain the task again.
- The tools you already use, connected over MCP.
- Clear control: how you steer an agent, and how you stop it.
- A way to review what an agent changed before you keep it.
- A choice of models, so each job can use the right one.
What is an AI workspace for coding agents?
It is a place to run Claude Code, Codex, Cursor and other coding agents against your repository, see what each is doing and review what they change. Coding work needs a real repository, visible changes and review, which a general chat workspace does not give you.
Friday is that, free and MIT-licensed, on your own computer. It runs 16 coding tools by name plus any ACP agent, side by side, on your own plans and sign-ins. Agents pick tasks from a shared board, message each other, and can each work in their own git worktree. You review each run’s changes line by line, with git built in. There is no account. See the docs or GitHub.
How do you choose an AI workspace?
- Name the work. Writing, coding, support and research need different things.
- Check control and review. Can you steer and stop agents, and read what they changed?
- Check the tools and models it connects to.
- Check where it runs and what it stores. A desktop app that runs on your computer keeps your repository there.
- Try it on one real task before you commit to it.