> For the complete documentation index, see [llms.txt](https://docs.duplocloud.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.duplocloud.com/docs/quick-start/install-the-devkit.md).

# Step 1: Install the DevKit

Download the DuploCloud DevKit and start building agents on your own machine.

The DevKit runs the DuploCloud platform on your own machine in containers. You use it to connect your cloud accounts, ask the AI agent questions about them, and build your own agents, which load into the running platform without a restart.

The DevKit is public on GitHub at [duplocloud/devkit](https://github.com/duplocloud/devkit). Its docs are the full, step-by-step guide. This page gives you the overview and links to the right place for each step.

## What you need

* A container runtime with Compose v2 (Docker or podman).
* Python 3.
* Access to an LLM: an Anthropic API key, AWS Bedrock credentials, or an Anthropic-compatible LLM gateway.
* An email address you can read right away. Work or personal addresses are fine. Privacy-relay and disposable addresses are not accepted.
* A few free local ports.

The details, including podman on Apple Silicon, are in [Prerequisites](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/prerequisites.md). On Windows, set up WSL 2 first: [Windows: set up WSL](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/windows-wsl.md).

## What happens on first run

You clone the repository and run its start script. The commands are in [Install and sign in](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/install.md).

1. **You enter an admin email.** This becomes your login for the local portal.
2. **DuploCloud requests a trial license for that email** and sends you a verification link. The run waits for you to click it, then continues on its own. If the wait times out, click the link anyway and run the script again; it picks up the same request.
3. **You set a password and choose your LLM provider.** The script then starts the platform and creates a workspace for you to build in.
4. **You sign in** to the portal at `http://localhost:4210`.
5. **You land on Get your first agent.** While no agents are installed, the portal opens on this page instead of an empty ticket list. It offers the same paths as this Quick Start: **Install an agent** from the Marketplace, **Create your own agent** with the DevKit or Agent Studio, or **Move to your own cloud**. Click **Skip for now** or **Don't show this again** to go straight to the portal.

{% hint style="warning" %}
DuploCloud issues **one license per email address**. The DevKit reuses or recovers the license it already has, so you shouldn't need another. If you already hold a license token, pass it with `--license <jwt>` to skip the trial request.
{% endhint %}

How the DevKit handles licenses is in [Configuration: Licensing](https://github.com/duplocloud/devkit/blob/main/docs/configuration.md#licensing). For how licenses work across DuploCloud, see [License](https://docs.duplocloud.com/docs/concepts/console/licensing.md).

## Where the DevKit stores things

Everything stays on your machine, in the folder you cloned:

| What                                                                | Where                                                              |
| ------------------------------------------------------------------- | ------------------------------------------------------------------ |
| Settings, credentials and your license token                        | The `.env` file                                                    |
| Your agents                                                         | `extensions/<name>/`. Upgrades and resets never touch this folder. |
| Platform data: database, file store, search index and loaded agents | Container volumes                                                  |

Stopping the DevKit keeps your data. Some reset commands delete the volumes but keep your license. [What each command destroys](https://github.com/duplocloud/devkit/blob/main/docs/cli-reference.md#what-each-command-destroys) lists exactly what each one removes.

## Where to go next

**Next step in the Quick Start:** [Try a Marketplace Agent](https://docs.duplocloud.com/docs/quick-start/try-a-marketplace-agent.md).

To connect your accounts and build in the DevKit:

| To do this                                        | Read                                                                                                                                                                                                                                          |
| ------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Connect an AWS account and ask the agent about it | [Connect AWS](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/connect-aws.md)                                                                                                                                             |
| Connect Azure, GCP, or another cloud or tool      | [Connecting Providers](https://docs.duplocloud.com/docs/concepts/providers/integrating-providers.md)                                                                                                                                          |
| Connect a Kubernetes cluster                      | [Connect Kubernetes](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/connect-kubernetes.md)                                                                                                                               |
| Build your first agent                            | [Build Your Own Agent](https://docs.duplocloud.com/docs/quick-start/build-your-own-agent.md), or the DevKit guide [Build your first agent](https://github.com/duplocloud/devkit/blob/main/docs/getting-started/build-your-first-extension.md) |
| Get a working setup fast                          | [Quickstart](https://github.com/duplocloud/devkit/blob/main/docs/quickstart.md)                                                                                                                                                               |
| Upgrade to a newer DevKit                         | [Upgrading](https://github.com/duplocloud/devkit/blob/main/docs/upgrading.md)                                                                                                                                                                 |
| Fix a problem                                     | [Troubleshooting](https://github.com/duplocloud/devkit/blob/main/docs/troubleshooting.md) and the [FAQ](https://github.com/duplocloud/devkit/blob/main/docs/faq.md)                                                                           |
| Look up a command or flag                         | [CLI reference](https://github.com/duplocloud/devkit/blob/main/docs/cli-reference.md)                                                                                                                                                         |

When your team is ready to use your agents, see [Deploy to Your Cloud](https://docs.duplocloud.com/docs/deploy-to-your-cloud/deploy-to-your-cloud.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.duplocloud.com/docs/quick-start/install-the-devkit.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
