Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 38 additions & 1 deletion content/agents/claude-code/quickstart.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,8 @@ import {DocsCTA} from '~/components/blog/CTA'

This guide will help you get started with Depot's remote agent sandboxes for running Claude Code agents. With Depot, you can run your coding agents in a secure, isolated environment in the cloud, allowing you to move your coding sessions off of your local machine and into fast remote environments where you can easily launch, resume, and share coding sessions.

The `depot claude` command runs in remote sandboxes by default, providing better isolation and collaboration capabilities.

## 1. Install the `depot` CLI

Before you can start running remote agents, you need to install the Depot CLI. The CLI is available for Mac and Linux.
Expand Down Expand Up @@ -135,4 +137,39 @@ This command immediately tells the Depot control plane to start a new agent sand

![Remote Claude Code session inside of Depot](/images/docs/remote-claude-code-session-ui.webp)

And that's it! You now have your Depot organization set up to make use of Depot agent sandboxes for Claude Code. You can pass different git repositories that your git credentials or Depot Code app have access to, you can specify different branches, you can resume a session using the `--resume` flag, and you can fork new sessions from an existing file systems using the `--session-id` flag and `--resume` flag at the same time.
And that's it! You now have your Depot organization set up to make use of Depot agent sandboxes for Claude Code.

## Remote vs Local execution

By default, `depot claude` runs in a remote sandbox environment. However, you can also choose to run locally if needed:

### Remote execution (default)

All commands run in the cloud with full isolation:

```shell
# Start a new remote session
depot claude --session-id my-feature

# Resume a remote session
depot claude --resume my-feature

# Work with a repository in remote sandbox
depot claude --repository https://github.com/user/repo.git --session-id repo-work
```

### Local execution

Run Claude locally while still leveraging Depot's session management:

```shell
# Run locally while saving session to Depot
depot claude --local --session-id local-work

# Resume a local session
depot claude --local --resume local-work
```

**Note:** Local execution only persists session information to Depot but doesn't provide the isolation and collaboration benefits of remote sandboxes.

You can pass different git repositories that your git credentials or Depot Code app have access to, you can specify different branches, you can resume a session using the `--resume` flag, and you can fork new sessions from an existing file systems using the `--session-id` flag and `--resume` flag at the same time.
14 changes: 9 additions & 5 deletions content/cli/reference.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -276,25 +276,29 @@ depot cache reset --project 12345678910

Run Claude Code in remote agent sandboxes backed by Depot with automatic session & file system saving and resuming. Sessions are stored by Depot and can be resumed by session ID, allowing you to collaborate on any session in your organization across any environment.

By default, Claude Code runs in a remote sandbox environment.
By default, `depot claude` runs in a remote sandbox environment, providing better isolation, reproducibility, and collaboration capabilities.

Note: All flags not recognized by `depot` are passed directly through to the Claude CLI. This includes Claude flags like `-p`, `--model`, etc.

**Example**
**Examples**

Start a new Claude Code session with a custom ID:
#### Remote execution (default behavior)

Start a new Claude Code session in a remote sandbox:

```shell
depot claude --session-id feature-auth-redesign
```

Resume an existing session:
Resume an existing remote session:

```shell
depot claude --resume feature-auth-redesign
```

Run Claude Code locally instead of in a sandbox:
#### Local execution

Run Claude Code locally instead of in a remote sandbox:

Note: This will only persist the Claude Code session information up to Depot, but not execute in a remote sandbox.

Expand Down