Prompt stage

The Prompt stage prompts the user to enter information. The submitted values are then stored in the flow context.

Overview

Use this stage to collect user input, such as profile attributes, passwords, invitation tokens, file uploads, or approval choices.

Submitted values are written to prompt_data in the flow context and can be consumed by later stages or policies.

Configuration options

Stage options

• Fields: the ordered set of prompt definitions included in the stage.
• Validation policies: optional policies evaluated after submission to validate the entered data.

Prompt field types

Each prompt field can use one of these types:

Type	Description
Text	Arbitrary text input.
Text (read-only)	Text displayed but not editable.
Text area	Multiline text input.
Text area (read-only)	Multiline text displayed but not editable.
Username	Text input with username uniqueness checks.
Email	Text input validated as an email address.
Password	Masked input. Multiple password fields in the same stage must match.
Number	Numeric input.
Checkbox	Boolean checkbox input.
Radio Button Group	Single-choice selection rendered as radio buttons.
Dropdown	Single-choice selection rendered as a dropdown.
Date	Date picker.
Date-time	Date and time picker.
File	File upload stored in flow context as a data URI.
Separator	Static separator element.
Hidden	Hidden value inserted into the form submission.
Static	Static value displayed as-is.
authentik: Locale	Locale selector populated from authentik-supported locales.

Prompt field attributes

Each prompt field definition includes these core attributes:

• field_key: key used to store the field value in prompt_data.
• label: label shown to the user.
• required: whether the field must be provided.
• placeholder: placeholder text, or for choice fields, the available choices.
• initial value: pre-filled value, or for choice fields, the default selected choice.
• order: numeric ordering among the prompt fields.
• Interpret placeholder as expression: evaluate the placeholder dynamically.
• Interpret initial value as expression: evaluate the initial value dynamically.

For Radio Button Group and Dropdown, the placeholder defines the valid choices. It can be a plain string, a list of values, or a list of objects with label and value.

Some field types also have special behavior:

• Username checks for uniqueness.
• Password fields within the same stage must all match.
• Hidden and Static use their configured values without user editing.
• Radio Button Group and Dropdown only allow values from the defined choice set.

Flow integration

Use this stage anywhere a flow needs user-provided input.

Common follow-ups include:

• A User Write stage to persist collected values
• An Email or Invitation stage to act on collected data
• Policy checks that read from request.context["prompt_data"]

Notes

Accessing submitted data

Prompt values are stored in prompt_data:

request.context.get("prompt_data", {}).get("<field_key>")

Dynamic choices and defaults

For choice fields, placeholder and initial value can be generated dynamically with expressions. The prompt context is available during evaluation.

Choice fields also support context overrides:

• <field_key>__choices overrides the available choices
• <field_key> overrides the initial value

For Radio Button Group and Dropdown, choices can be plain values or objects like:

return [
    "first option",
    42,
    {"label": "another option", "value": "some value"},
]

For fixed-choice fields, the default value must be one of the valid choices.

When prompt expressions are evaluated, they run in the same general expression environment used by authentik mappings and policies and also have access to prompt_context.

Validation

Additional validation can be done with validation policies bound directly to the Prompt stage. For example, to ensure two password fields match:

if request.context["prompt_data"]["password"] == request.context["prompt_data"]["password_repeat"]:
    return True

ak_message("Passwords don't match.")
return False