Recommended releases

Download Released

The AI Field Automator module lets you attach AI generation rules directly
to Backdrop field instances. When enabled on a field, the module can populate
that field from another field, from rendered entity content, from tokens, from
media inputs, or from search results depending on the selected automator type.

This module is intended for the Backdrop ai ecosystem and follows the
same provider/model configuration used by the rest of the AI modules in
this codebase.

Features

  • Per-field AI automator configuration in Field UI
  • Multiple worker modes: direct, batch, queue/cron, and widget action
  • Rule ordering by weight
  • Text, scalar, reference, media, structured, search, and domain-specific
    automator types
  • Optional async widget-action generation when ai_async is enabled

Requirements

  • Backdrop CMS 1.x
  • PHP 8.2 or later
  • The ai module

Some automator types require additional modules or field types, for example:

  • token support for token-mode prompts
  • ai_search for vector-search automators
  • media/file/image-related modules for media automators

Installation

  1. Place the module in modules/custom/
  2. Enable it with bee pm-enable ai_field_automator or from
    admin/modules
  3. Make sure at least one AI provider/model is configured

Configuration

  1. Go to a field settings form, for example:
    admin/structure/types/manage/[content-type]/fields/[field]/edit
  2. Open the AI Automator fieldset
  3. Enable the automator for that field
  4. Choose an automator type
  5. Configure the prompt, base field, model, worker type, and any
    type-specific settings
  6. Save the field

The module also provides an overview page at:

  • /admin/config/ai/ai-field-automator

Worker Modes

  • Direct runs during save
  • Batch runs via Backdrop batch processing after form submit
  • Queue/Cron queues work for cron
  • Widget action adds a Generate with AI button to supported field widgets

Automator Types

The module includes a broad set of automator types for:

  • Text fields
  • Numeric, boolean, email, telephone, link, and option fields
  • Taxonomy and entity references
  • Images, audio, speech, and video workflows
  • JSON, metatag, moderation-state, address, office-hours, FAQ, and chart data
  • Search API AI vector search and Views extraction

The full type reference is documented in:

More Documentation

Implementation details and module architecture are documented separately in:

Focused Xdebug notes for the post.field_summary widget-action automator are in:

Notes

  • Configuration is stored per field, not as standalone config entities
  • The module creates an ai_automator_status field on bundles that have active
    automators
  • For queue/cron workers, choose an explicit text format when needed because
    cron may run as anonymous

Issues

Bugs and feature requests should be reported in the Issue Queue.

Current Maintainer

Justin Keiser

Credits

License

This project is GPL v2 software. See the LICENSE.txt file in this directory for complete text.