From db9945ad0141f181e59f66b82dd72140b4f89b1c Mon Sep 17 00:00:00 2001 From: David Karlsson <35727626+dvdksn@users.noreply.github.com> Date: Mon, 12 Oct 2026 19:50:15 +0000 Subject: [PATCH 1/4] docs: make DAP getting started a complete first session The Console getting started page ended at sandbox launch, leaving readers to assemble task execution, saving results, and lifecycle instructions from other pages. Add a sample project review, result inspection and saving, optional cloud SSH editor access, and pause or delete guidance. Preserve the sandbox options reference and existing heading anchors. --- .../manuals/agentic-platform/get-started.md | 175 ++++++++++++++++-- 1 file changed, 156 insertions(+), 19 deletions(-) diff --git a/content/manuals/agentic-platform/get-started.md b/content/manuals/agentic-platform/get-started.md index adc6c42269ab..0e1fb50ae6b2 100644 --- a/content/manuals/agentic-platform/get-started.md +++ b/content/manuals/agentic-platform/get-started.md @@ -1,43 +1,180 @@ --- title: Get started with Docker Agentic Platform linkTitle: Get started -description: Choose a kit and start a Docker Agentic Platform sandbox. -keywords: docker agentic platform, get started, kits, agents, sandbox, live terminal +description: Run your first cloud sandbox, ask an agent to review a sample project, save the result, and pause or delete the sandbox when you're finished. +keywords: docker agentic platform, get started, cloud sandboxes, agents, ssh, vscode, cursor weight: 10 aliases: - /agentic-platform/concepts/platform-model/ --- -Start a sandbox from the Console to work with an agent in the cloud. +Use a cloud sandbox to have an agent review a project without setting up the +project on your computer. In this walkthrough, you'll launch Claude Code, +clone a public sample application, and save a project review. Then you'll +pause the sandbox to return later or delete it when you're finished. + +You can complete the walkthrough in the browser. An optional step shows how +to open the same project in VS Code or Cursor over SSH. ## Before you begin -You need a personal Docker account with an active -[Docker Agentic Platform subscription](signup.md#activate-cloud-access). +You need: -If your kit needs a model provider API key, add it under **Secrets** or enter -it when you launch the sandbox. +- A personal Docker account with an active + [Docker Agentic Platform subscription](signup.md#activate-cloud-access). +- An Anthropic API key for the Claude Code kit. Add it under **Secrets** or + enter it when you launch the sandbox. Your model provider bills inference + separately from Docker's sandbox compute charges. -Browse [Kits](kits.md) to choose an agent such as Claude Code, Codex, or Hermes, -or use Shell to work without a pre-installed agent. The credentials you need -depend on the kit. For example, Hermes needs an Anthropic or OpenAI API key. +This example uses a public repository, so you don't need a GitHub token. +For other agents and their credentials, see [Kits](kits.md) and +[Secrets](secrets.md). ## Start a sandbox 1. Open [Docker Agentic Platform](https://agentic-platform.docker.com/) and select **New**. -2. Choose a kit. -3. Add any credentials requested by the launcher. -4. Review the [sandbox options](#sandbox-options), including the selected - network policies, and adjust them for your task. -5. Select **Run**. If a required credential is missing, the launcher opens its +2. Choose the Claude Code kit. +3. Add your Anthropic API key if it isn't already saved. Don't enter it in + the agent's prompt or in a file inside the sandbox. +4. Review the selected network policies. The sandbox needs access to + `github.com:443` to clone the sample, along with the model provider access + supplied by the kit. If GitHub isn't allowed, follow + [Create a policy](policies.md#create-a-policy) to add an allow rule and + select that policy in the launcher. An explicit deny rule takes precedence + over an allow rule. +5. Keep the default compute size and **auto** platform. Under **Run for**, + choose **1 hour**, and under **When the time is up**, choose **Stop** to + keep the sandbox's files when the timer expires. You don't need MCP tools + for this task. +6. Select **Run**. If a required credential is missing, the launcher opens its input so you can add it. -When the sandbox is ready, its terminal opens. Use it to work with your agent, -or run commands if you chose Shell. After launch, you can't change the selected -credentials, tools, network policies, or compute size. +When the sandbox is ready, its terminal opens with Claude Code. Complete any +agent setup prompts. After launch, you can't change the selected credentials, +tools, network policies, or compute size. See [Sandbox options](#sandbox-options) +for other launch settings. + +## Give the agent a task + +Your local files aren't mounted in a cloud sandbox. Clone the sample project +inside it so the agent has source code to work with. + +Select **+** in the terminal tab bar to open another shell in the same sandbox. +Run: + +```console +$ git clone https://github.com/docker/welcome-to-docker.git /home/agent/workspace/project +``` + +Return to the Claude Code terminal tab and enter this prompt: + +```text +Review the application in /home/agent/workspace/project. +Read the source, package.json, and Dockerfile. Don't modify the application +or install dependencies. +Write /home/agent/workspace/review.md with: +- A short explanation of what the application does. +- The commands defined in package.json for running and building it. +- An explanation of how the Dockerfile builds and serves the application. +- One small improvement you'd recommend, with the relevant file path. +Base the review on the files you read. Don't claim to have run the application. +``` + +Review any permission requests from the agent and approve the actions needed +for the task. The agent's response can vary; the result to look for is +`/home/agent/workspace/review.md`. + +## Inspect and save the result + +When the agent finishes, switch to the shell tab and read the review: + +```console +$ cat /home/agent/workspace/review.md +$ cat /home/agent/workspace/project/package.json +$ cat /home/agent/workspace/project/Dockerfile +``` + +Compare the commands and Dockerfile explanation with the source files. If +something is missing or incorrect, return to the agent tab and ask it to +correct the review. + +Copy the review's text from the terminal into a file on your computer. Files +don't sync automatically with your computer or a remote repository. Saving a +copy means you can keep the result after deleting the sandbox. + +For larger files or directories, use the CLI to +[transfer files](/manuals/ai/sandboxes/cloud/usage.md#transfer-files). +For your own projects, you can also commit and push changes to a repository; +include a [GitHub credential](secrets.md#github-credential) when launching a +sandbox that needs to push changes. + +## Open the project in your editor (optional) + +Use SSH to inspect the project and review in VS Code or Cursor. Your editor +runs on your computer, while the files and terminal commands stay in the +cloud sandbox. + +1. Install the [Docker Sandboxes CLI](/manuals/ai/sandboxes/install.md), + version 0.45.0 or later. +2. In a terminal on your computer, sign in with the same Docker account you + used in the Console and configure cloud SSH access: + + ```console + $ sbx login + $ sbx --cloud setup ssh + $ sbx --cloud ls + ``` + +3. Find your sandbox in the list and copy its `sbx_`-prefixed ID. Test the + connection, replacing `` with that ID: + + ```console + $ ssh @sbx_cloud + ``` + + Run `exit` to return to your computer's terminal. +4. In VS Code, install the + [Remote - SSH extension](https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-ssh). + In Cursor, install its Remote - SSH support. +5. Open your editor's Command Palette and run **Remote-SSH: Connect to Host**. + Enter `@sbx_cloud`, using the same ID as in the SSH command. +6. Open `/home/agent/workspace` in the remote window. You'll find the cloned + application in `project` and the agent's output in `review.md`. + +The first connection installs the editor's server inside the sandbox. If the +connection reports a blocked download, check the sandbox's +[network policies](policies.md) for the host shown in the error. + +You can also copy the review to your computer from a local terminal: + +```console +$ sbx --cloud cp :/home/agent/workspace/review.md ./review.md +``` + +## Pause, keep running, or delete the sandbox + +Closing the browser or disconnecting your editor leaves the sandbox running. +An agent can continue working until you pause or delete the sandbox, or its +timer expires. Compute is billed while the sandbox runs, even when the agent +has finished its task. + +Return to **Sandboxes** in the Console and choose what to do next: + +- Pause the sandbox to stop compute charges and keep its files. Confirm that + it is paused in the list. To continue later, reopen the sandbox and resume + it. Resuming starts a fresh timer with the original duration. +- Leave it running if the agent still has work to do. Check the expiration + time: the **Stop** action you selected ends the session when the timer + expires, including background processes. +- Delete it when you no longer need it. Save the review or push any changes + you want to keep first. Deletion removes files stored only in the sandbox. -Return to **Sandboxes** to reopen the terminal. +For more detail, see [Lifecycle actions](sandboxes.md#manage-the-lifecycle). +To run a task from your own terminal, follow the +[cloud CLI walkthrough](/manuals/ai/sandboxes/cloud/_index.md#get-started). +For programmatic workflows, see the +[Sandboxes API quickstart](/manuals/ai/sandboxes-api/get-started.md). ## Sandbox options From ad82980b48555796effd4f32e016c1f97f44db69 Mon Sep 17 00:00:00 2001 From: David Karlsson <35727626+dvdksn@users.noreply.github.com> Date: Mon, 12 Oct 2026 19:55:41 +0000 Subject: [PATCH 2/4] docs: simplify the first sandbox launch The walkthrough narrated default settings and launcher behavior that readers do not need to act on. Use the default launch settings, keep custom network policy guidance to a short aside, and describe the default timeout in the lifecycle step. --- .../manuals/agentic-platform/get-started.md | 21 +++++++------------ 1 file changed, 7 insertions(+), 14 deletions(-) diff --git a/content/manuals/agentic-platform/get-started.md b/content/manuals/agentic-platform/get-started.md index 0e1fb50ae6b2..123955f3e87b 100644 --- a/content/manuals/agentic-platform/get-started.md +++ b/content/manuals/agentic-platform/get-started.md @@ -37,18 +37,11 @@ For other agents and their credentials, see [Kits](kits.md) and 2. Choose the Claude Code kit. 3. Add your Anthropic API key if it isn't already saved. Don't enter it in the agent's prompt or in a file inside the sandbox. -4. Review the selected network policies. The sandbox needs access to - `github.com:443` to clone the sample, along with the model provider access - supplied by the kit. If GitHub isn't allowed, follow - [Create a policy](policies.md#create-a-policy) to add an allow rule and - select that policy in the launcher. An explicit deny rule takes precedence - over an allow rule. -5. Keep the default compute size and **auto** platform. Under **Run for**, - choose **1 hour**, and under **When the time is up**, choose **Stop** to - keep the sandbox's files when the timer expires. You don't need MCP tools - for this task. -6. Select **Run**. If a required credential is missing, the launcher opens its - input so you can add it. +4. Keep the defaults and select **Run**. + +On your first launch, the default **Open** network policy permits outbound +access to any host. If you're using a custom policy, make sure it permits +GitHub and your model provider. See [Network policies](policies.md). When the sandbox is ready, its terminal opens with Claude Code. Complete any agent setup prompts. After launch, you can't change the selected credentials, @@ -165,8 +158,8 @@ Return to **Sandboxes** in the Console and choose what to do next: it is paused in the list. To continue later, reopen the sandbox and resume it. Resuming starts a fresh timer with the original duration. - Leave it running if the agent still has work to do. Check the expiration - time: the **Stop** action you selected ends the session when the timer - expires, including background processes. + time: the default **Stop** action stops the sandbox after one hour, including + background processes, and keeps its files. - Delete it when you no longer need it. Save the review or push any changes you want to keep first. Deletion removes files stored only in the sandbox. From 405cee88becc666f71a54bf7f09cb0177b3d0429 Mon Sep 17 00:00:00 2001 From: David Karlsson <35727626+dvdksn@users.noreply.github.com> Date: Tue, 13 Oct 2026 16:17:34 +0000 Subject: [PATCH 3/4] docs: tighten the cloud sandbox walkthrough The walkthrough added shell-based inspection and manual copying steps that distracted from working with the agent. Have the agent clone the sample and summarize its review in the conversation. Replace the inspection ceremony with concise file persistence guidance and retain CLI transfer and Git as ways to save work. --- .../manuals/agentic-platform/get-started.md | 68 +++++++------------ 1 file changed, 23 insertions(+), 45 deletions(-) diff --git a/content/manuals/agentic-platform/get-started.md b/content/manuals/agentic-platform/get-started.md index 123955f3e87b..29a5c27afa13 100644 --- a/content/manuals/agentic-platform/get-started.md +++ b/content/manuals/agentic-platform/get-started.md @@ -43,27 +43,18 @@ On your first launch, the default **Open** network policy permits outbound access to any host. If you're using a custom policy, make sure it permits GitHub and your model provider. See [Network policies](policies.md). -When the sandbox is ready, its terminal opens with Claude Code. Complete any -agent setup prompts. After launch, you can't change the selected credentials, -tools, network policies, or compute size. See [Sandbox options](#sandbox-options) -for other launch settings. +When the sandbox is ready, its terminal opens with Claude Code. After launch, +you can't change the selected credentials, tools, network policies, or compute +size. See [Sandbox options](#sandbox-options) for other launch settings. ## Give the agent a task -Your local files aren't mounted in a cloud sandbox. Clone the sample project -inside it so the agent has source code to work with. - -Select **+** in the terminal tab bar to open another shell in the same sandbox. -Run: - -```console -$ git clone https://github.com/docker/welcome-to-docker.git /home/agent/workspace/project -``` - -Return to the Claude Code terminal tab and enter this prompt: +Your local files aren't mounted in a cloud sandbox. Ask Claude Code to clone +and review a public sample project: ```text -Review the application in /home/agent/workspace/project. +Clone https://github.com/docker/welcome-to-docker.git into +/home/agent/workspace/project, then review the application. Read the source, package.json, and Dockerfile. Don't modify the application or install dependencies. Write /home/agent/workspace/review.md with: @@ -72,39 +63,28 @@ Write /home/agent/workspace/review.md with: - An explanation of how the Dockerfile builds and serves the application. - One small improvement you'd recommend, with the relevant file path. Base the review on the files you read. Don't claim to have run the application. +Summarize your findings in your response. ``` -Review any permission requests from the agent and approve the actions needed -for the task. The agent's response can vary; the result to look for is -`/home/agent/workspace/review.md`. - -## Inspect and save the result - -When the agent finishes, switch to the shell tab and read the review: - -```console -$ cat /home/agent/workspace/review.md -$ cat /home/agent/workspace/project/package.json -$ cat /home/agent/workspace/project/Dockerfile -``` +Claude Code summarizes the project in the conversation and saves the review +at `/home/agent/workspace/review.md`. You can ask follow-up questions or give +it another task in the same session. -Compare the commands and Dockerfile explanation with the source files. If -something is missing or incorrect, return to the agent tab and ask it to -correct the review. +## Keep your work -Copy the review's text from the terminal into a file on your computer. Files -don't sync automatically with your computer or a remote repository. Saving a -copy means you can keep the result after deleting the sandbox. +The cloned project and review are stored in the sandbox. Files don't sync +automatically with your computer or a remote repository. They remain available +when you pause and resume the sandbox, but are lost when you delete it. -For larger files or directories, use the CLI to -[transfer files](/manuals/ai/sandboxes/cloud/usage.md#transfer-files). -For your own projects, you can also commit and push changes to a repository; -include a [GitHub credential](secrets.md#github-credential) when launching a -sandbox that needs to push changes. +To keep files outside the sandbox, use the CLI to +[transfer files](/manuals/ai/sandboxes/cloud/usage.md#transfer-files), or commit +and push changes to a repository. Include a +[GitHub credential](secrets.md#github-credential) when launching a sandbox that +needs to push changes. ## Open the project in your editor (optional) -Use SSH to inspect the project and review in VS Code or Cursor. Your editor +Open the project and review in VS Code or Cursor over SSH. Your editor runs on your computer, while the files and terminal commands stay in the cloud sandbox. @@ -135,11 +115,9 @@ cloud sandbox. 6. Open `/home/agent/workspace` in the remote window. You'll find the cloned application in `project` and the agent's output in `review.md`. -The first connection installs the editor's server inside the sandbox. If the -connection reports a blocked download, check the sandbox's -[network policies](policies.md) for the host shown in the error. +The first connection installs the editor's server inside the sandbox. -You can also copy the review to your computer from a local terminal: +To download the review, run this command in a terminal on your computer: ```console $ sbx --cloud cp :/home/agent/workspace/review.md ./review.md From 55613a74ebc4a7904ea687f327cf387882b84ceb Mon Sep 17 00:00:00 2001 From: David Karlsson <35727626+dvdksn@users.noreply.github.com> Date: Tue, 13 Oct 2026 16:17:57 +0000 Subject: [PATCH 4/4] docs: explain a successful cloud SSH connection The SSH step told readers to exit without explaining what a successful connection opens. State that SSH opens a shell inside the cloud sandbox before returning to the local terminal. --- content/manuals/agentic-platform/get-started.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/content/manuals/agentic-platform/get-started.md b/content/manuals/agentic-platform/get-started.md index 29a5c27afa13..45ee64d51c51 100644 --- a/content/manuals/agentic-platform/get-started.md +++ b/content/manuals/agentic-platform/get-started.md @@ -106,7 +106,8 @@ cloud sandbox. $ ssh @sbx_cloud ``` - Run `exit` to return to your computer's terminal. + A successful connection opens a shell inside the cloud sandbox. Run `exit` + to return to your computer's terminal. 4. In VS Code, install the [Remote - SSH extension](https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-ssh). In Cursor, install its Remote - SSH support.