SSD Nodes Learn Hosting plans →
Guides Matt ConnorBy Matt Connor

Claude Cowork vs Claude Code: how to choose

Cowork runs in an Anthropic sandbox and touches folders you connect. Claude Code runs on your machine, a VPS or Anthropic's cloud. Checked October 2026.

What is the difference between Claude Cowork and Claude Code?

Claude Cowork and Claude Code are two Anthropic agents built for different material. Cowork works on documents and office tools, inside folders and connectors you grant it. Claude Code works on a codebase and a shell, either on the machine where you start it, which can be your laptop or a VPS (virtual private server) you rent, or in a cloud session on Anthropic-managed infrastructure.

The short routing rule: if the deliverable is a file a colleague will open, use Cowork. If the deliverable is a commit or a running service, use Claude Code.

Every capability claim below was re-checked on 27 October 2026 against Anthropic's own pages: the Cowork product page, the Cowork architecture overview, the Cowork getting started guide, the Claude Code documentation and the Claude Code cloud sessions page. Availability, surfaces and plan gating are all moving, so check the date on any comparison you read, including this one.

Where does each one run the work?

This is the difference that drives all the others.

Cowork has two execution modes. As checked on 27 October 2026, the architecture page described them separately. In a local session, the agent loop ran natively on the device. That covered conversation handling, file reads and writes in connected folders, web fetches, and local plugin MCP servers. Code execution happened inside an isolated Linux virtual machine on that same device, with network egress filtering and syscall restrictions applied to it.

Local sessions are no longer the default. On Pro, Max and Ultra, only tasks started on the device before 6 October 2026 stay local. Previously created scheduled tasks move to the cloud automatically. Every new Pro, Max and Ultra task runs in the cloud.

In a cloud session, the agent's work and code execution run in an isolated, temporary environment on Anthropic-managed infrastructure. Each session gets its own sandbox, created when the session starts and destroyed when it ends. As checked on 27 October 2026, sandboxes shared no state with each other, could not reach private, internal, link-local or cloud-metadata addresses, and sent every outbound connection through a proxy they could not reconfigure or bypass, allowing only listed destinations.

The payoff of the cloud session is that the work survives you walking away: close the laptop and the task keeps going. The cost is that the sandbox is not your machine. It has no direct access to your home or company network.

Claude Code is the other shape. The command line interface (CLI) runs as an ordinary process on a machine you choose, so it inherits that machine's filesystem, its network position and its credentials.

curl -fsSL https://claude.ai/install.sh | bash
cd your-project
claude

That installer covers macOS, Linux and WSL. Claude Code also runs natively on Windows: from PowerShell the install line is irm https://claude.ai/install.ps1 | iex, and CMD and WinGet installers exist too. You are prompted to log in on first use. Run those same lines inside an SSH session and Claude Code is now running on the server, with your repository and your systemd units in reach.

Claude Code also has a cloud mode, which makes the comparison less tidy and more honest. claude --cloud "fix the failing auth test" starts a session on an isolated Anthropic-managed virtual machine, and that machine clones your GitHub remote at your current branch rather than your local working copy. claude --teleport pulls the session back into your terminal with its branch and history. Cloud sessions, which in the browser are still called Claude Code on the web, are available on Pro, Max, Ultra and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats. Enterprise users governed by custom roles also need the cloud sessions permission to start sessions on Anthropic's infrastructure. Cloud sessions can also run on your organization's self-hosted infrastructure when routed there.

What can each one touch?

Cowork touches what you connect, and nothing else. On the desktop app you grant access to specific folders, and Claude can read, write and permanently delete files inside them. Anthropic's safety guidance is direct about the consequence: make "a dedicated working folder for Claude", and keep folders holding credentials or personal records out of it.

The product page names connectors for Microsoft 365, Slack and Amplitude, and its sample tasks read from and save to folders in Drive. That is the office surface, and it is where Cowork earns its place.

A cloud Cowork session reaches your local files through the desktop app on that device. Three things follow. The reach stops at the folders that member connected. From web or mobile, local files are reachable only for a session started on desktop or for a scheduled task. And "if the desktop app is offline, a session in the cloud can't reach the device", so a scheduled overnight task that needs a local spreadsheet finds nothing when the laptop is shut. Work that lives entirely in cloud connectors keeps running.

Claude Code touches whatever the user account running it can touch. That means the repository, and also apt, systemctl, docker, your database socket, and any host reachable from the network the server sits in. Nobody grants that folder by folder, which is the point and also the risk.

The tool that narrows it is the sandboxed Bash tool, turned on with /sandbox inside a session. It ships with Claude Code and it uses the operating system to enforce a filesystem and network boundary on sandboxed shell commands and their child processes, so Claude runs most commands without stopping to ask. Excluded commands and approved unsandboxed retries run outside that boundary. It works on macOS, Linux and WSL2. Claude Code runs on native Windows too, but there it runs commands unsandboxed, so to use the sandbox on a Windows machine you run Claude Code inside a WSL2 distribution.

How do you steer a task and review the result?

Cowork is built around handing off a task and checking what comes back. Existing Dispatch users assign work through one ongoing thread they can reach from their phone or their desktop, and a push notification arrives when the task is done or when Claude needs their go-ahead. Claude sends you the outcome, a spreadsheet or a memo, instead of showing you every step. As of 17 November 2026, Dispatch is in limited beta for some Pro, Max and Ultra plans and not open to new users, and it needs your computer awake with the app open. New users start and steer tasks through Cowork in the cloud. Progress indicators show what it is doing at each step when you want to follow along.

In the earlier Cowork experience, approvals come in three settings, and the names say what they do. Manual pauses and asks you to allow or deny each action. Auto lets Claude keep working while a safety check reviews each action and blocks unsafe ones. Skip means Claude does not pause, except for connector tools set to "Always ask", and nothing checks its actions automatically, so the help centre says to use it only when you completely trust every action in the task. "Always ask" is not fully supported in Dispatch or research yet. The new unified Claude experience offers Auto and Manual. The help centre says the plain thing, and it is worth repeating: no mode replaces your judgement, and you remain responsible for what the agent does on your behalf.

Claude Code is built around a diff. You watch it work in the terminal, you read the change it proposes, and review happens in the workflow you already have: branches and pull requests. When you want to leave the desk, Remote Control exposes a running local session for monitoring from a browser or the mobile app, which is the same pattern as steering a Claude Code session from your phone.

Scheduling exists on both sides. Cowork schedules tasks that run in the cloud, so they no longer need your computer to be awake. Claude Code splits it: cloud scheduled tasks, formerly called routines, run in the cloud, and desktop scheduled tasks run on your machine with direct access to local files.

Do connectors and plugins replace MCP servers and skills?

No, because both products stand on the same extension stack. MCP (model context protocol) is an open standard for connecting an agent to outside tools and data. Skills are packaged instructions the agent loads when they become relevant. Plugins bundle skills and connectors together, and in Cowork you manage them under Customize.

The real difference is where the server runs. Remote MCP connectors are available across Claude, Cowork and Claude Desktop. Local MCP servers are the exception: "plugins that include local MCP servers work through the desktop app only", so a Cowork session on the web or on your phone needs the desktop app to use them.

Claude Code takes the same MCP servers and adds what a repository needs. CLAUDE.md at the project root is read at the start of every session. Hooks run shell commands before or after an action, so a formatter can run on every edit. Subagents split work into separate context windows. Those files live in the repo, so they get reviewed and versioned like code.

If you already run your own MCP servers on a VPS, the remote ones are reachable from both products, and that is a way to give Cowork access to something Anthropic does not ship a connector for. The skills layer is shared too, so a skill you write is portable, and agent skills covers the format.

What leaves your machine?

Answer this before you connect a folder.

In a Cowork cloud session, a lot leaves: "the agent's work, including any local files it opens through the desktop app, is processed on Anthropic's servers rather than staying on the device". That is the design, and it is what lets the session survive a closed laptop. As checked on 27 October 2026, a local Cowork session kept the agent loop and the file handling on the device instead. On Pro, Max and Ultra, local sessions now cover only tasks started on the device before 6 October 2026; scheduled tasks migrate to the cloud automatically.

With Claude Code running locally, the process and the files stay on your machine or your server, and so does the command output. What travels is the context: the file contents and command output the model is asked to reason about, because the model itself runs on Anthropic's servers. Nothing about the word "local" changes that part. claude --cloud changes more, since the cloud session clones your repository into an Anthropic-managed virtual machine. When the repository has no git remote, or is on github.com without the Claude GitHub App installed on it, Claude Code instead bundles your local repository, including uncommitted changes to tracked files, and uploads it. On macOS, Linux and WSL, it leaves out uncommitted changes to files named like credentials or keys and files managed by git filters; on native Windows, those changes upload as they are.

Which one should you use?

Use Cowork when the work is documents and office systems. Building a deck from a folder of research. Reconciling two spreadsheets. Drafting a briefing from Slack threads and email. Turning a folder of PDFs into a summary someone will read. These tasks run many steps, they cross several apps, and the deliverable is a file.

Use Claude Code when the work is a codebase, a terminal or a server. Fixing a failing test. Bisecting a regression. Writing a migration and running it. Working out why nginx returns 502 after a deploy. Anything where success is checked by a command you run and an output you read.

The tie-breaker is that check. If you verify the result by opening a document, Cowork fits. If you verify it by running a command, use Claude Code, because it can run the command itself and read what came back.

Cowork runs on desktop on every paid plan (Pro, Max, Ultra, Team and Enterprise). The web (claude.ai) and the Claude mobile apps are listed for Pro, Max, Ultra and Team, and for Enterprise where an admin has enabled them. Those sessions run in the cloud. Anthropic's unified Claude experience, where chat and Cowork share one home and agentic work runs in the cloud, is rolling out to Pro, Max and Ultra, and is available on Team and Enterprise where Chat and Cowork unified is enabled. Since 16 November 2026, eligible Enterprise organizations without an explicit admin choice default to enabled when both Chat and Cloud code execution and file creation are on; an owner's explicit choice to enable or disable it stays in place. For organizations still on by default, eligible custom roles receive it on 8 December 2026, even if an admin previously disabled it for that role. Cowork in the cloud is still labelled beta on Team. A Claude in Chrome side panel is a further surface, listed for all paid plans, and on Enterprise an owner has to enable both Claude in Chrome and Cowork in the cloud. The product page's FAQ no longer labels web or mobile as beta: it says Cowork "runs on desktop, with web and mobile", and its answer on assigning tasks from your phone lists Pro, Max, Ultra and Team automatically and lets Enterprise opt in. The help centre's getting started guide still words the Team beta as covering web and mobile. The desktop app ships for macOS and Windows, and Anthropic's Linux desktop app, itself in beta for Ubuntu and Debian, includes the Cowork tab. Plan gating moves faster than anything else here, so read the current page rather than a write-up. Carry one operational note with you: Cowork consumes more usage than standard chat, because an agent that runs many steps spends many turns doing it.

If your real question is which coding agent to adopt, rather than which of these two, the comparison against Cursor, Codex and Copilot is the more useful page.

What changes when the work lives on a VPS

A Cowork cloud sandbox cannot reach your home or company network, so it cannot directly open an SSH connection to a server on that network, read /var/log/nginx/error.log there or restart a unit. A local Cowork session, where you still have one, runs on your own device and so sits closer to your network, with access through connected folders, local MCP servers and desktop tools.

Claude Code on the server has none of those limits, which is why the safety work moves onto you. Run it as a normal user with sudo rules you chose, not as root. Keep the session inside a project directory. Turn on /sandbox so Bash commands are held to a boundary the kernel enforces. Keep credentials out of the working tree, because anything the agent reads becomes context it can repeat back. Running Claude Code safely on a VPS covers the account setup, the firewall and the persistent terminal session in detail.

The pattern that works: the server holds the repository and the services, Claude Code runs there in a terminal session that outlives your SSH connection, and you attach from wherever you are. Work continues while your laptop sleeps because the process lives on a machine that does not sleep. That is the same benefit a Cowork cloud session gives you, reached a different way, on hardware you control.

The failures you should expect

A scheduled Cowork task did nothing overnight. The desktop app was closed. A cloud session keeps running when the app goes offline, but it cannot reach the device, so any step that reads a connected folder finds nothing to read. Move the input into a connected cloud service, or leave the machine awake with the app open.

Cowork read a web page and then did something you never asked for. That is prompt injection: instructions hidden in external content that Claude reads as part of a legitimate task. The defence is a smaller blast radius. Connect one working folder instead of your home directory, and be careful with unfamiliar MCP servers and plugins.

Claude Code touched a file you did not expect. A local session has the reach of the user account that started it. Read the diff before you commit, and work on a branch you can throw away.

A cloud Claude Code session could not see your latest work. When your repository is on GitHub with the Claude GitHub App installed, claude --cloud clones the remote at your current branch, not your local checkout, so uncommitted changes are invisible to it. Push first, then start the session.

FAQ

Can Claude Cowork edit code and run commands like Claude Code?

It can run code, since code execution happens in the sandbox in a cloud session. As checked on 27 October 2026, local sessions executed code in an isolated Linux virtual machine. It is not built around a repository. Cowork also supports hooks and subagents through plugins. For a one-off script over a spreadsheet, Cowork is fine. For work in a codebase you maintain, Claude Code is the tool, because the review layer is git and the success check is a command.

Can Claude Cowork reach my VPS or my internal server?

The cloud sandbox cannot directly reach your home or company network. Network access follows the configured egress settings. If you want an agent working on a server, install Claude Code on that server and run it there.

Do I need the Claude Desktop app to use Cowork?

For local files and local MCP servers, yes. Cowork runs on desktop on every paid plan, and on web and mobile on Pro, Max, Ultra and Team, and on Enterprise where an admin has enabled it. Running Cowork in the cloud is still labelled beta on Team, and the unified Claude experience is rolling out to Pro, Max and Ultra and is available on Team and Enterprise through admin settings. A cloud session can read and write files in connected folders only while the desktop app is open on that computer, and from web or mobile only for a session started on desktop or a scheduled task. Browser actions in the Claude in Chrome side panel do not require Desktop: Claude can read, click, type, navigate and fill forms there. Anthropic's Chrome guide says driving Chrome from another surface still needs Desktop open and connected. Computer use and plugins with local MCP servers need the desktop app, and so do live artifacts created before 19 August 2026. Cowork also has a built-in browser in Desktop, available from web or mobile while the desktop app is online. A task that uses only cloud connectors such as Google Drive or Slack runs without the app.

Can I reuse the same skills and MCP servers in both?

Mostly. Remote MCP connectors work across Claude, Cowork and Claude Desktop, and skills are shared, so a lot of your setup is portable between them. Local MCP servers are the exception in Cowork, because they work through the desktop app only. Claude Code then adds repository-level configuration such as CLAUDE.md; hooks and subagents are available in both.

Was this article useful?