> For the complete documentation index, see [llms.txt](https://docs.bloxchain.app/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.bloxchain.app/guides/ways-to-configure.md).

# Ways to configure an account

There is more than one way to set up a Bloxchain smart account. Pick the route that matches who is doing the work and how often you will repeat it.

**Where:** Discover, inside an account, or your own tooling · **You need:** a wallet with a role that may sign the change

## The five routes

Every route ends in the same place: a signed change that the account contract checks against its own policy. The chain does not care which one you used.

| Route                               | Choose this when                                                       | Signs the change                      |
| ----------------------------------- | ---------------------------------------------------------------------- | ------------------------------------- |
| [Presets](#presets)                 | You are starting from nothing and want an account in one transaction   | Your wallet, at deploy                |
| [The Platform UI](#the-platform-ui) | You are making a handful of changes and want to see what each one does | Your wallet, per batch                |
| [Signed files](#signed-files)       | The wallet that decides is not on the machine that broadcasts          | The approver, then a broadcaster      |
| [The SDK](#the-sdk)                 | You are configuring many accounts, or the setup must be repeatable     | Your script's key, or a wallet client |
| [Agent-assisted](#agent-assisted)   | You want a draft prepared for you, and you will review it              | The agent, then a person broadcasts   |

## Presets

**Discover** lists first-party presets: Personal security, Treasury, Asset issuance. Choosing one deploys a governed account in a single transaction.

Today every preset deploys the same governed account, and it lands as a **Vault**: a finished, locked account with no spending capability yet ([what that means](/platform-guide/policies/actions.md#what-a-new-account-already-has)). Capabilities are added afterwards through one of the routes below. A preset is a starting point, not a finished configuration.

Trade-off: fastest to a working account, but it does not wire roles or permissions for you.

See [Discover presets](/platform-guide/discover.md).

## The Platform UI

Inside an account, **Policies** holds Security, Actions, Gates and the allowlists, and **On-chain access** holds Signers and Roles. You change one thing at a time, the Platform batches the change, and you sign it.

Trade-off: the clearest route, because each screen tells you what is missing (**Flow incomplete**, gate gaps, empty allowlists). It is also the slowest if you have twenty accounts to set up.

See [Policies](/platform-guide/policies.md) and [On-chain access](/platform-guide/access.md).

## Signed files

A signed change can be exported as a `.bloxmeta.json` file, carried to another device, imported and broadcast there. This covers an air-gapped signer, a hardware wallet on a separate laptop, and a team where the person who decides is not the person who pays gas.

Trade-off: no relay and no sign-in are needed, but the file is a bearer instrument and it expires.

See [Signed files](/platform-guide/operate/signed-files.md) and [Approve from another device](/guides/cross-device-approval.md).

## The SDK

The Protocol's TypeScript SDK does everything the Platform does: register actions, create roles, assign permissions per step, grant gates, manage allowlists. A script can configure a new account the same way every time, and run in CI.

Trade-off: repeatable and reviewable in source control, but you own the key handling and the error paths.

See [Configure programmatically](/developers/sdk-setup/configure-programmatically.md).

## Agent-assisted

An AI agent holding a narrowly permitted signer key can draft and sign a request or a configuration change. You import the file, read the decoded preview and broadcast it. The agent never holds the broadcaster key or the owner key, and the on-chain policy still constrains everything it signs.

Trade-off: useful for drafting and for repetitive work, and a person still decides. More AI tooling is in development.

See [Agent-assisted requests](/guides/agent-assisted-requests.md).

## Mixing routes

Most accounts use more than one. A common pattern is a preset to deploy, the Platform UI for Security and the first actions, then the SDK or an agent for the repetitive parts, with signed files whenever the approving wallet is somewhere else.

## Related

* [Design roles and permissions](/guides/roles-and-permissions.md)
* [Organization governance](/guides/organization-governance.md)
* [Limits and defaults](/resources/limits-and-defaults.md)
* [Help and setup checks](/platform-guide/help.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.bloxchain.app/guides/ways-to-configure.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.
