centersl/docker-docs
1
0
Fork 0
mirror of https://github.com/docker/docs.git synced 2026-04-12 14:25:46 +07:00
Code Issues Packages Projects Releases Wiki Activity

ai: Disambiguate Docker Agent, Gordon, and Docker Sandboxes

Browse Source 
Add AI overview landing page with decision table and relationship overview.
Reframe Gordon as "AI-powered assistant" instead of "AI agent" to avoid
confusion with Docker Agent. Rename sandbox agent page from cagent.md to
docker-agent.md. Add cross-references where products overlap. Fix doubled
"Docker Docker Agent" in agentic AI guide.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: David Karlsson <35727626+dvdksn@users.noreply.github.com>
This commit is contained in:
David Karlsson
2026-03-06 09:57:48 +01:00
parent e91de6298b
commit 0b0568a5be
8 changed files with 101 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
content/guides/agentic-ai.md
View File

@@ -12,8 +12,8 @@ params:
> [!TIP]
>
> This guide uses the familiar Docker Compose workflow to orchestrate agentic AI
> applications. For a smoother development experience, check out [Docker
> Docker Agent](../manuals/ai/docker-agent/_index.md), a purpose-built agent runtime that
> applications. For a smoother development experience, check out
> [Docker Agent](../manuals/ai/docker-agent/_index.md), a purpose-built agent runtime that
> simplifies running and managing AI agents.
## Introduction
@@ -65,11 +65,11 @@ all works together.
To follow this guide, you need to:
- [Install Docker Desktop 4.43 or later](../get-started/get-docker.md)
- [Enable Docker Model Runner](/manuals/ai/model-runner.md#enable-dmr-in-docker-desktop)
- At least the following hardware specifications:
- VRAM: 3.5 GB
- Storage: 2.31 GB
- [Install Docker Desktop 4.43 or later](../get-started/get-docker.md)
- [Enable Docker Model Runner](/manuals/ai/model-runner.md#enable-dmr-in-docker-desktop)
- At least the following hardware specifications:
- VRAM: 3.5 GB
- Storage: 2.31 GB
## Step 1: Clone the sample application
@@ -90,8 +90,9 @@ run in the cloud. This particular example uses the [Gemma 3 4B
model](https://hub.docker.com/r/ai/gemma3) with a context size of `10000`.
Hardware requirements:
- VRAM: 3.5 GB
- Storage: 2.31 GB
- VRAM: 3.5 GB
- Storage: 2.31 GB
If your machine exceeds those requirements, consider running the application with a larger
context size or a larger model to improve the agents performance. You can easily
@@ -113,7 +114,7 @@ To run the application locally, follow these steps:
incorrect fact in the prompt and hit enter. An agent searches DuckDuckGo to
verify it and another agent revises the output.
![Screenshot of the application](./images/agentic-ai-app.png)
![Screenshot of the application](./images/agentic-ai-app.png)
3. Press ctrl-c in the terminal to stop the application when you're done.
@@ -136,7 +137,7 @@ services:
depends_on:
- mcp-gateway
models:
gemma3 :
gemma3:
endpoint_var: MODEL_RUNNER_URL
model_var: MODEL_RUNNER_MODEL
@@ -160,16 +161,16 @@ models:
The app consists of three main components:
- The `adk` service, which is the web application that runs the agentic AI
application. This service talks to the MCP gateway and model.
- The `mcp-gateway` service, which is the MCP gateway that connects the app
to external tools and services.
- The `models` block, which defines the model to use with the application.
- The `adk` service, which is the web application that runs the agentic AI
application. This service talks to the MCP gateway and model.
- The `mcp-gateway` service, which is the MCP gateway that connects the app
to external tools and services.
- The `models` block, which defines the model to use with the application.
When you examine the `compose.yaml` file, you'll notice two notable elements for the model:
- A service‑level `models` block in the `adk` service
- A top-level `models` block
- A service‑level `models` block in the `adk` service
- A top-level `models` block
These two blocks together let Docker Compose automatically start and connect
your ADK web app to the specified LLM.
@@ -189,7 +190,7 @@ example, it uses the [`duckduckgo` MCP
server](https://hub.docker.com/mcp/server/duckduckgo/overview) to perform web
searches.
> [!TIP]
> [!TIP]
>
> Looking for more MCP servers to use? Check out the [Docker MCP
> Catalog](https://hub.docker.com/catalogs/mcp/).

46
content/manuals/ai-overview.md Normal file
View File

@@ -0,0 +1,46 @@
---
title: Docker AI overview
linkTitle: Overview
description: Docker's AI tools help you build, run, and manage AI-powered applications and workflows.
weight: 1
keywords: docker, ai, gordon, docker agent, sandboxes, model runner, mcp
params:
sidebar:
group: AI
---
Docker provides tools for working with AI across your development workflow.
Each tool serves a different purpose.
## Which tool do I need?
| I want to... | Use | CLI command |
| --------------------------------------------------------------- | -------------------------------------------------------- | ---------------- |
| Get AI help with Docker tasks (containers, images, Dockerfiles) | [Gordon](./ai/gordon/) | `docker ai` |
| Run AI models locally with an OpenAI-compatible API | [Model Runner](./ai/model-runner/) | `docker model` |
| Connect AI tools to external services via MCP | [MCP Catalog and Toolkit](./ai/mcp-catalog-and-toolkit/) | `docker mcp` |
| Build and orchestrate custom multi-agent teams | [Docker Agent](./ai/docker-agent/) | `docker agent` |
| Run coding agents in isolated environments | [Docker Sandboxes](./ai/sandboxes/) | `docker sandbox` |
## How these tools relate
**Gordon** is Docker's built-in AI assistant. It helps with Docker-specific
tasks like debugging containers, writing Dockerfiles, and managing images. You
interact with it through Docker Desktop or the `docker ai` command.
**Docker Agent** is an open-source framework for defining teams of AI agents
in YAML. You configure agents with specific roles, models, and tools, then
run them from your terminal. Docker Agent is a general-purpose agent runtime,
not specific to Docker tasks.
**Docker Sandboxes** provides isolated microVM environments for running coding
agents. It supports multiple agents including Claude Code, Codex, Copilot,
Gemini, and Docker Agent. Sandboxes is the isolation layer — the agents
themselves are separate tools.
**Model Runner** lets you run LLMs locally. Other tools like Docker Agent can
use Model Runner as a model provider.
**MCP Catalog and Toolkit** manages connections between AI tools and external
services using the Model Context Protocol. Gordon, Docker Agent, and
third-party tools can all use MCP servers configured through the Toolkit.

13
content/manuals/ai/docker-agent/_index.md
View File

@@ -16,10 +16,15 @@ keywords: [ai, agent, docker agent, cagent]
{{< summary-bar feature_name="Docker Agent" >}}
[Docker Agent](https://github.com/docker/cagent) is an open source tool for building
teams of specialized AI agents. Instead of prompting one generalist model, you
define agents with specific roles and instructions that collaborate to solve
problems. Run these agent teams from your terminal using any LLM provider.
[Docker Agent](https://github.com/docker/cagent) is an open-source framework
for building teams of specialized AI agents. Instead of prompting one
generalist model, you define agents with specific roles and instructions that
collaborate to solve problems. Run these agent teams from your terminal using
any LLM provider.
> [!NOTE]
> Docker Agent is a framework for building and running custom agent teams.
> For Docker's built-in AI assistant, see [Gordon](/ai/gordon/) (`docker ai`).
## Why agent teams

5
content/manuals/ai/gordon/_index.md
View File

@@ -1,6 +1,7 @@
---
title: Gordon
description: AI agent for Docker workflows - execute tasks, debug issues, and
description:
AI assistant for Docker workflows - execute tasks, debug issues, and
manage containers with intelligent assistance
weight: 40
params:
@@ -15,7 +16,7 @@ aliases:
{{< summary-bar feature_name="Gordon" >}}
Gordon is an AI agent that takes action on your Docker workflows. It analyzes
Gordon is an AI-powered assistant that takes action on your Docker workflows. It analyzes
your environment, proposes solutions, and executes commands with your
permission. Available in Docker Desktop and via the `docker ai` CLI command.

4
content/manuals/ai/gordon/concepts/capabilities.md
View File

@@ -14,13 +14,13 @@ explains what Gordon can do and the tools it uses.
Gordon uses five capabilities to take action on your behalf:
- Specialized agents for specific Docker tasks
- Specialized sub-agents for specific Docker tasks
- Shell access to run commands
- Filesystem access to read and write files
- Knowledge base of Docker documentation and best practices
- Web access to fetch external resources
## Agent architecture
## Internal architecture
Gordon uses a primary agent that handles most tasks, with a specialized
sub-agent for specific workflows:

4
content/manuals/ai/sandboxes/_index.md
View File

@@ -1,6 +1,6 @@
---
title: Docker Sandboxes
description: Run AI agents in isolated environments
description: Run AI coding agents in isolated environments
weight: 20
params:
sidebar:
@@ -91,7 +91,7 @@ Docker Sandboxes works with multiple AI coding agents:
- **Copilot** - GitHub Copilot agent (in development)
- **Gemini** - Google's Gemini agent (in development)
- **OpenCode** - Multi-provider agent with TUI interface (in development)
- **Docker Agent** - Docker's multi-provider coding agent (in development)
- **[Docker Agent](/ai/docker-agent/)** - Docker's multi-provider coding agent (in development)
- **Kiro** - Interactive agent with device flow auth (in development)
- **Shell** - Minimal sandbox for manual agent installation

22
content/manuals/ai/sandboxes/agents/_index.md
View File

@@ -12,16 +12,16 @@ inside microVMs with private Docker daemons.
## Supported agents
| Agent | Command | Status | Notes |
| ----------- | ---------- | ------------ | ----------------------------------------- |
| Claude Code | `claude` | Experimental | Most tested implementation |
| Codex | `codex` | Experimental | In development |
| Copilot | `copilot` | Experimental | In development |
| Gemini | `gemini` | Experimental | In development |
| Docker Agent | `cagent` | Experimental | In development |
| Kiro | `kiro` | Experimental | In development |
| OpenCode | `opencode` | Experimental | In development |
| Custom shell | `shell` | Experimental | Minimal environment for manual setup |
| Agent | Command | Status | Notes |
| --------------------------------- | ---------- | ------------ | ------------------------------------ |
| Claude Code | `claude` | Experimental | Most tested implementation |
| Codex | `codex` | Experimental | In development |
| Copilot | `copilot` | Experimental | In development |
| Gemini | `gemini` | Experimental | In development |
| [Docker Agent](/ai/docker-agent/) | `cagent` | Experimental | Also available as a standalone tool |
| Kiro | `kiro` | Experimental | In development |
| OpenCode | `opencode` | Experimental | In development |
| Custom shell | `shell` | Experimental | Minimal environment for manual setup |
## Experimental status
@@ -67,7 +67,7 @@ agent (no fallback authentication methods are used).
See the agent-specific documentation:
- [Claude Code](./claude-code.md)
- [cagent](./cagent.md)
- [Docker Agent](./docker-agent.md)
- [Codex](./codex.md)
- [Copilot](./copilot.md)
- [Gemini](./gemini.md)

12
content/manuals/ai/sandboxes/agents/cagent.md → content/manuals/ai/sandboxes/agents/docker-agent.md
View File

@@ -3,15 +3,19 @@ title: Docker Agent sandbox
description: |
Use Docker Agent in Docker Sandboxes with multi-provider authentication
supporting OpenAI, Anthropic, and more.
keywords: docker, sandboxes, docker agent, ai agent, multi-provider, authentication
keywords: docker, sandboxes, docker agent, multi-provider, authentication
aliases:
- /ai/sandboxes/agents/cagent/
- /manuals/ai/sandboxes/agents/cagent/
weight: 60
---
{{< summary-bar feature_name="Docker Sandboxes" >}}
This guide covers authentication, configuration, and usage of Docker Agent in
a sandboxed environment. [Docker Agent](/ai/docker-agent/) is Docker's open source coding
agent that supports multiple providers.
This page covers running [Docker Agent](/ai/docker-agent/) inside Docker
Sandboxes. Docker Agent is also available as a standalone CLI tool. See the
full documentation for standalone usage, configuration reference, and building
agent teams.
## Quick start