<!--
This file was generated by Spark. Do not edit it by hand.
-->
# DSL: Ash.Flow.Dsl



## flow
Details about the flow itself, like description and the successful return type.


### Nested DSLs
 * [argument](#flow-argument)





### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`api`](#flow-api){: #flow-api } | `module` |  | An api to use by default when calling actions |
| [`description`](#flow-description){: #flow-description } | `String.t` |  | A description of the flow |
| [`trace_name`](#flow-trace_name){: #flow-trace_name } | `String.t` |  | The name to use when creating traces. Defaults to the short name. |
| [`short_name`](#flow-short_name){: #flow-short_name } | `atom` |  | A short name to use for the flow. Defaults to the last to parts of the module name, underscored. |
| [`returns`](#flow-returns){: #flow-returns } | `any` |  | The step or step that should constitute the return value. |



## flow.argument
```elixir
argument name, type
```


An argument to be passed into the flow




### Examples
```
argument :params, :map do
  default %{}
end

```

```
argument :retries, :integer do
  allow_nil? false
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#flow-argument-name){: #flow-argument-name .spark-required} | `atom` |  | The name to use for the argument |
| [`type`](#flow-argument-type){: #flow-argument-type .spark-required} | `module` |  | The type of the argument. See `Ash.Type` for more. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`default`](#flow-argument-default){: #flow-argument-default } | `(-> any) \| mfa \| any` |  | A default value to use for the argument if not provided |
| [`allow_nil?`](#flow-argument-allow_nil?){: #flow-argument-allow_nil? } | `boolean` | `true` | Whether or not the argument value may be nil |
| [`constraints`](#flow-argument-constraints){: #flow-argument-constraints } | `keyword` | `[]` | Constraints to provide to the type when casting the value. See the type's documentation for more information. |





### Introspection

Target: `Ash.Flow.Argument`




## steps
The steps to run.


### Nested DSLs
 * [map](#steps-map)

 * [branch](#steps-branch)

 * [transaction](#steps-transaction)

 * [create](#steps-create)
 * [debug](#steps-debug)
 * [update](#steps-update)
 * [destroy](#steps-destroy)
 * [validate](#steps-validate)
 * [read](#steps-read)
 * [run_flow](#steps-run_flow)
 * [custom](#steps-custom)


### Examples
```
steps do
  # invokes a create action
  create :create_post, MyApp.Post, :create
end

```




## steps.map
```elixir
map name, over
```


Runs a set of steps for each item in a provided list.




### Examples
```
map :create_users, range(1, arg(:count)) do
  output :create_user

  create :create_user, Org, :create do
    input %{
      first_name: {Faker.Person, :first_name, []},
      last_name: {Faker.Person, :last_name, []}
    }
  end
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-map-name){: #steps-map-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`over`](#steps-map-over){: #steps-map-over } | `any` |  | The value to be iterated over. Will be available inside the `map` step as `element(:map_step_name)` |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`output`](#steps-map-output){: #steps-map-output } | `atom` |  | Which step to use when constructing the output list. Defaults to the last step. |
| [`short_name`](#steps-map-short_name){: #steps-map-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-map-wait_for){: #steps-map-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-map-touches_resources){: #steps-map-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-map-halt_if){: #steps-map-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-map-halt_reason){: #steps-map-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-map-description){: #steps-map-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.Map`

## steps.branch
```elixir
branch name, condition
```


Runs a set of steps based on a given value.




### Examples
```
branch :create_users, result(:create_users?) do
  output :create_user

  create :create_user, Org, :create do
    input %{
      first_name: {Faker.Person, :first_name, []},
      last_name: {Faker.Person, :last_name, []}
    }
  end
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-branch-name){: #steps-branch-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`condition`](#steps-branch-condition){: #steps-branch-condition } | `any` |  | A template that must evaluate to `true` for the branch to be executed. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`output`](#steps-branch-output){: #steps-branch-output } | `atom` |  | Which step to use as the output. Defaults to the last step. |
| [`short_name`](#steps-branch-short_name){: #steps-branch-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-branch-wait_for){: #steps-branch-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-branch-touches_resources){: #steps-branch-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-branch-halt_if){: #steps-branch-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-branch-halt_reason){: #steps-branch-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-branch-description){: #steps-branch-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.Branch`

## steps.transaction
```elixir
transaction name, resource
```


Runs a set of steps in a transaction.




### Examples
```
transaction :create_user_with_org do
  touches_resources [User, Org]

  create :create_user, User, :create do
    input %{
      first_name: {Faker.Person, :first_name, []},
      last_name: {Faker.Person, :last_name, []}
    }
  end

  create :create_org, Org, :create do
    input %{
      user_id: path(result(:create_user), :id),
      name: {Faker.Color, :name, []}
    }
  end
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-transaction-name){: #steps-transaction-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-transaction-resource){: #steps-transaction-resource } | `module \| list(module)` |  | The Ash resource to use for the transaction. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`output`](#steps-transaction-output){: #steps-transaction-output } | `any` |  | Which step or steps to use when constructing the output. Defaults to the last step. |
| [`timeout`](#steps-transaction-timeout){: #steps-transaction-timeout } | `timeout` |  | A timeout to apply to the transaction. |
| [`short_name`](#steps-transaction-short_name){: #steps-transaction-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-transaction-wait_for){: #steps-transaction-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-transaction-touches_resources){: #steps-transaction-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-transaction-halt_if){: #steps-transaction-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-transaction-halt_reason){: #steps-transaction-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-transaction-description){: #steps-transaction-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.Transaction`

## steps.create
```elixir
create name, resource, action
```


Declares a step that will call a create action on a resource.




### Examples
```
create :create_post, MyApp.Post, :create

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-create-name){: #steps-create-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-create-resource){: #steps-create-resource .spark-required} | `any` |  | The resource to call the action on. |
| [`action`](#steps-create-action){: #steps-create-action .spark-required} | `any` |  | The action to call on the resource. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`upsert?`](#steps-create-upsert?){: #steps-create-upsert? } | `boolean` | `false` | Whether or not this action is always an upsert. |
| [`upsert_identity`](#steps-create-upsert_identity){: #steps-create-upsert_identity } | `atom` |  | The identity to use for the upsert. |
| [`short_name`](#steps-create-short_name){: #steps-create-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-create-wait_for){: #steps-create-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-create-touches_resources){: #steps-create-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-create-halt_if){: #steps-create-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-create-halt_reason){: #steps-create-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-create-description){: #steps-create-description } | `String.t` |  | A description for the step. |
| [`api`](#steps-create-api){: #steps-create-api } | `any` |  | The api to use when calling the action. Defaults to the api set in the `flow` section. |
| [`tenant`](#steps-create-tenant){: #steps-create-tenant } | `any` |  | A tenant to use for the operation. May be a template or a literal value. |
| [`input`](#steps-create-input){: #steps-create-input } | `any` |  | A template for the input. |





### Introspection

Target: `Ash.Flow.Step.Create`

## steps.debug
```elixir
debug name
```


Declares a step that will inspect its input and provide
additional debug information.




### Examples
```
debug :show_some_information do
  input %{post: result(:create_post)}
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-debug-name){: #steps-debug-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`input`](#steps-debug-input){: #steps-debug-input } | `any` |  | A template for the input. |
| [`short_name`](#steps-debug-short_name){: #steps-debug-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-debug-wait_for){: #steps-debug-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`halt_if`](#steps-debug-halt_if){: #steps-debug-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-debug-halt_reason){: #steps-debug-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-debug-description){: #steps-debug-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.Debug`

## steps.update
```elixir
update name, resource, action
```


Declares a step that will call a update action on a resource.




### Examples
```
update :update_post, MyApp.Post, :update do
  record result(:get_post)
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-update-name){: #steps-update-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-update-resource){: #steps-update-resource .spark-required} | `any` |  | The resource to call the action on. |
| [`action`](#steps-update-action){: #steps-update-action .spark-required} | `any` |  | The action to call on the resource. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`record`](#steps-update-record){: #steps-update-record .spark-required} | `any` |  | The record to be updated, can use template helpers, e.g `result(:step_name)`. |
| [`short_name`](#steps-update-short_name){: #steps-update-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-update-wait_for){: #steps-update-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-update-touches_resources){: #steps-update-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-update-halt_if){: #steps-update-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-update-halt_reason){: #steps-update-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-update-description){: #steps-update-description } | `String.t` |  | A description for the step. |
| [`api`](#steps-update-api){: #steps-update-api } | `any` |  | The api to use when calling the action. Defaults to the api set in the `flow` section. |
| [`tenant`](#steps-update-tenant){: #steps-update-tenant } | `any` |  | A tenant to use for the operation. May be a template or a literal value. |
| [`input`](#steps-update-input){: #steps-update-input } | `any` |  | A template for the input. |





### Introspection

Target: `Ash.Flow.Step.Update`

## steps.destroy
```elixir
destroy name, resource, action
```


Declares a step that will call a destroy action on a resource.




### Examples
```
destroy :destroy_post, MyApp.Post, :destroy

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-destroy-name){: #steps-destroy-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-destroy-resource){: #steps-destroy-resource .spark-required} | `any` |  | The resource to call the action on. |
| [`action`](#steps-destroy-action){: #steps-destroy-action .spark-required} | `any` |  | The action to call on the resource. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`record`](#steps-destroy-record){: #steps-destroy-record .spark-required} | `any` |  | The record to be updated, can use template helpers, e.g `result(:step_name)`. |
| [`short_name`](#steps-destroy-short_name){: #steps-destroy-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-destroy-wait_for){: #steps-destroy-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-destroy-touches_resources){: #steps-destroy-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-destroy-halt_if){: #steps-destroy-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-destroy-halt_reason){: #steps-destroy-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-destroy-description){: #steps-destroy-description } | `String.t` |  | A description for the step. |
| [`api`](#steps-destroy-api){: #steps-destroy-api } | `any` |  | The api to use when calling the action. Defaults to the api set in the `flow` section. |
| [`tenant`](#steps-destroy-tenant){: #steps-destroy-tenant } | `any` |  | A tenant to use for the operation. May be a template or a literal value. |
| [`input`](#steps-destroy-input){: #steps-destroy-input } | `any` |  | A template for the input. |





### Introspection

Target: `Ash.Flow.Step.Destroy`

## steps.validate
```elixir
validate name, resource, action
```


Validates some input against an action.




### Examples
```
validate :update_post, MyApp.Post, :update do
  record result(:get_post)
  only_keys [:name]
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-validate-name){: #steps-validate-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-validate-resource){: #steps-validate-resource .spark-required} | `any` |  | The resource to call the action on. |
| [`action`](#steps-validate-action){: #steps-validate-action .spark-required} | `any` |  | The action to call on the resource. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`record`](#steps-validate-record){: #steps-validate-record } | `any` |  | The record to be created/updated/destroyed. If the value is `nil` and would be required by the action type, the step is skipped and `nil` is the result of the step. |
| [`only_keys`](#steps-validate-only_keys){: #steps-validate-only_keys } | `list(atom \| list(atom))` |  | A list of keys or paths to keys that should be validated. Others will be ignored, and errors generated for other fields will be ignored. |
| [`short_name`](#steps-validate-short_name){: #steps-validate-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-validate-wait_for){: #steps-validate-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-validate-touches_resources){: #steps-validate-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-validate-halt_if){: #steps-validate-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-validate-halt_reason){: #steps-validate-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-validate-description){: #steps-validate-description } | `String.t` |  | A description for the step. |
| [`api`](#steps-validate-api){: #steps-validate-api } | `any` |  | The api to use when calling the action. Defaults to the api set in the `flow` section. |
| [`tenant`](#steps-validate-tenant){: #steps-validate-tenant } | `any` |  | A tenant to use for the operation. May be a template or a literal value. |
| [`input`](#steps-validate-input){: #steps-validate-input } | `any` |  | A template for the input. |





### Introspection

Target: `Ash.Flow.Step.Update`

## steps.read
```elixir
read name, resource, action
```


Declares a step that will call a read action on a resource.




### Examples
```
read :destroy_post, MyApp.Post, :destroy

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-read-name){: #steps-read-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`resource`](#steps-read-resource){: #steps-read-resource .spark-required} | `any` |  | The resource to call the action on. |
| [`action`](#steps-read-action){: #steps-read-action .spark-required} | `any` |  | The action to call on the resource. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`get?`](#steps-read-get?){: #steps-read-get? } | `boolean` | `false` | Whether or not read action is expected to return a single result or `nil`. Set to `true` automatically if `get? true`. |
| [`not_found_error?`](#steps-read-not_found_error?){: #steps-read-not_found_error? } | `boolean` | `true` | Whether or not finding no record should result in a not found error |
| [`short_name`](#steps-read-short_name){: #steps-read-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-read-wait_for){: #steps-read-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-read-touches_resources){: #steps-read-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-read-halt_if){: #steps-read-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-read-halt_reason){: #steps-read-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-read-description){: #steps-read-description } | `String.t` |  | A description for the step. |
| [`api`](#steps-read-api){: #steps-read-api } | `any` |  | The api to use when calling the action. Defaults to the api set in the `flow` section. |
| [`tenant`](#steps-read-tenant){: #steps-read-tenant } | `any` |  | A tenant to use for the operation. May be a template or a literal value. |
| [`input`](#steps-read-input){: #steps-read-input } | `any` |  | A template for the input. |





### Introspection

Target: `Ash.Flow.Step.Read`

## steps.run_flow
```elixir
run_flow name, flow
```


Runs another flow as part of the current flow.
The return value of the step is the return value of the flow.




### Examples
```
run_flow :get_org, GetOrgByName do
  input %{
    name: arg(:org_name)
  }

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-run_flow-name){: #steps-run_flow-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`flow`](#steps-run_flow-flow){: #steps-run_flow-flow .spark-required} | `atom` |  | The flow to run. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`input`](#steps-run_flow-input){: #steps-run_flow-input } | `any` |  | A template for the input. |
| [`short_name`](#steps-run_flow-short_name){: #steps-run_flow-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-run_flow-wait_for){: #steps-run_flow-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-run_flow-touches_resources){: #steps-run_flow-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-run_flow-halt_if){: #steps-run_flow-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-run_flow-halt_reason){: #steps-run_flow-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-run_flow-description){: #steps-run_flow-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.RunFlow`

## steps.custom
```elixir
custom name, custom
```


Runs a custom step module.

See `Ash.Flow.Step` for the necessary callbacks and more information.




### Examples
```
custom :do_custom_thing, MyApp.DoCustomThing do
  input %{...}
end

```

```
custom :do_custom_thing, {MyApp.DoCustomThing, opt1: :foo, opt2: :bar} do
  input %{...}
end

```



### Arguments

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`name`](#steps-custom-name){: #steps-custom-name .spark-required} | `atom` |  | The name of the step. Will be used when expressing dependencies, and step inputs. |
| [`custom`](#steps-custom-custom){: #steps-custom-custom } | `(any, any -> any) \| module` |  | The module that implements the step behaviour. Also accepts a 2 argument function that takes the input and the context. |
### Options

| Name | Type | Default | Docs |
|------|------|---------|------|
| [`input`](#steps-custom-input){: #steps-custom-input } | `any` |  | A template for the input. |
| [`async?`](#steps-custom-async?){: #steps-custom-async? } | `boolean` | `false` | Whether or not this step can be run outside of the current process. |
| [`short_name`](#steps-custom-short_name){: #steps-custom-short_name } | `String.t` |  | Set a short name for the step. Will be used when building things like mermaid charts. |
| [`wait_for`](#steps-custom-wait_for){: #steps-custom-wait_for } | `any` |  | Ensures that the step happens after the configured step or steps. This is a template who's results are not used, only awaited. |
| [`touches_resources`](#steps-custom-touches_resources){: #steps-custom-touches_resources } | `list(atom)` |  | A list of resources touched by any custom logic in this step. This is used in the case that this step is run in a transaction. This is primarily only needed for `custom` steps. |
| [`halt_if`](#steps-custom-halt_if){: #steps-custom-halt_if } | `any` |  | Halts the step by emitting an error (with an `Ash.Error.Flow.Halted`). Can use template variables. See the section on Halting for more. |
| [`halt_reason`](#steps-custom-halt_reason){: #steps-custom-halt_reason } | `any` | `:halted` | Configures the reason for the `halt_if` clause. |
| [`description`](#steps-custom-description){: #steps-custom-description } | `String.t` |  | A description for the step. |





### Introspection

Target: `Ash.Flow.Step.Custom`





<style type="text/css">.spark-required::after { content: "*"; color: red !important; }</style>