Fields

Fields are the specific values within a parameter. While parameters define categories like “Temperament” or “Class”, fields define the individual options within those categories — “Confused”, “Happy”, “CS-180”, and so on.

What is a Field?

A field is a single value that belongs to a parameter. Fields are the building blocks that get attached to documents and scenarios, letting you tag simulation content with precise characteristics.

University example: The “Temperament” parameter has four fields:

Other examples include “CS-180” and “CS-251” as fields under the “Class” parameter, or “Almost Empty (1)” through “Hectic (10)” as fields under the “Crowdedness” parameter.

How It Connects: The 5-Step Workflow

Fields contribute to step 2 of the Glow workflow through their role in organizing documents:

1. Create Personas — Define AI student characters with temperaments and behaviors.
2. Assign to Scenarios — Attach personas and documents to scenarios. Documents are tagged with fields to indicate their specific characteristics.
3. Add to Simulations — Group scenarios into simulations.
4. Add to Cohorts — Assign profiles and simulations to a cohort.
5. Run Attempts — Learners launch simulations; the AI uses field-tagged content during the conversation.

Fields connect to parameters (each field belongs to a parameter) and to documents (documents are tagged with fields for categorization and filtering).

Create a field

Via the CLI

Calls below use $GLOW_INSTANCE_URL + $GLOW_TOKEN — see Authentication to export them once.

glow fields create --body '{
  "fields": [
    {
      "name": "Confused",
      "description": "The AI student character is uncertain and struggles to follow along"
    }
  ]
}'

Via the API

All endpoints use POST and require an Authorization: Bearer header.

curl -X POST $GLOW_INSTANCE_URL/field/create \
  -H "Authorization: Bearer $GLOW_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "fields": [
      {
        "name": "CS-180",
        "description": "Introduction to Computer Science course field"
      }
    ]
  }'

Conditional parameters and multiple descriptions

Conditional Parameters

Fields can have conditional parameters. This means a field can specify that when it is selected, certain additional parameters become relevant. This enables dynamic scenario configuration where selecting one field value unlocks further options.

Example: Selecting the “CS-180” field under the “Class” parameter might conditionally require the “Location” parameter, since CS-180 has specific classroom and lab locations.

The API exposes this through conditional_parameter_ids on the field draft, and the get endpoint returns a conditional_parameters section.

Descriptions

Fields support multiple descriptions. This is useful when a field needs different levels of detail for different contexts — a short label for list views and a longer description for the AI to use during simulation.

The get endpoint accepts a descriptions_search parameter, letting you search through a field’s descriptions when it has many.

Drafts

Fields support the draft workflow. Changes are saved as a draft before being published, with optimistic concurrency via expected_version.

# Save a field draft
glow fields draft --body '{
  "name": "Happy",
  "description": "The AI student character is positive, engaged, and receptive to instruction"
}'

Via the API, use PATCH /field/draft with fields like input_draft_id, expected_version, name, description, flag_id, department_ids, and conditional_parameter_ids.

Filtering and search

Fields support rich search and filtering:

• Parameter — Find fields belonging to specific parameters (parameter_ids, parameter_search)
• Persona — Find fields used by specific personas (persona_ids, persona_search)
• Department — Scope to a department (filter_department_ids, department_search)

# Search fields by parameter
curl -X POST $GLOW_INSTANCE_URL/field/search \
  -H "Authorization: Bearer $GLOW_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "parameter_search": "Temperament",
    "page_size": 20
  }'

Bulk operations

Bulk delete and update follow the canonical all-matching shape — pass either explicit IDs, or all: true with flat filter fields + optional excluded_ids + a patch for updates.

# Delete every field under a retired parameter, except a manual keeplist
glow fields delete --body '{
  "all": true,
  "parameter_ids": ["retired-parameter-uuid"],
  "excluded_ids": ["field-keep-canonical"]
}'
 
# Bulk update via patch
glow fields update --body '{
  "all": true,
  "filter_department_ids": ["dept-archive"],
  "patch": { "archived": true }
}'

See the Personas guide for the canonical write-up of this pattern.

Common Operations

Related

• Fields API
• Fields CLI
• Parameters Guide — the categories that contain fields
• Documents Guide — documents reference fields
• Departments Guide — scope fields to departments