yifanfeng97/hyper-extract/record-designer

Record Designer: model/list/set

Input from Brainstorm

Receive design specs from brainstorm:

• What fields to extract
• Field types and requirements
• Deduplication needs (for set)

Workflow

1. Confirm type (model/list/set)
2. Design fields (output.fields)
3. Configure identifiers (identifiers.item_id for set)
4. Set display (display.label)
5. Write guideline (guideline)
6. Review and output YAML

Critical: Output vs Guideline

Key Principle: Schema defines "WHAT", Guideline defines "HOW TO DO WELL". DO NOT repeat schema definitions in guideline.

Type Confirmation

Output Template

language: en
 
name: [TemplateName]
type: [model/list/set]
tags: [...]
description: '...'
 
output:
  description: '...'
  fields:
    - name: field_name
      type: str/int/float/list
      description: '...'
      required: true/false
      default: '...'
 
guideline:
  target: 'You are a [domain] expert...'
  rules: [...]
 
identifiers: {}  # Or item_id for set
 
display:
  label: '{field_name}'

Type-Specific Output

model

type: model
identifiers: {}  # Not needed

list

type: list
identifiers: {}  # Not needed

set

type: set
identifiers:
  item_id: [deduplication field]

Cases by Type

Important: Load only the case matching user's selected type.

Reference Files

Important: Check these files only when needed for the specific design task.

Design Checklist

• [ ] All fields have clear semantic meaning?
• [ ] Field types are appropriate (str/int/float/list)?
• [ ] Required vs optional is reasonable?
• [ ] Default values are safe/meaningful?
• [ ] For set: item_id can uniquely identify records?
• [ ] Display label references correct fields?

Guideline (Must Check)

• [ ] No field definitions repeated from schema?
• [ ] Extraction strategy defined?
• [ ] Quality requirements specified?
• [ ] Common mistakes warned?
• [ ] For set: deduplication rules clear?

Multi-language

• [ ] zh descriptions use pure Chinese
• [ ] en descriptions use pure English