What This Looks Like
The AI returns structured output, a form-like response, JSON, table rows, metadata, or a generated record, but required fields are missing. The output may parse correctly and may even look organized, yet a validator, reviewer, importer, or downstream workflow cannot proceed because expected fields are absent.
Why It Matters
Missing required fields create incomplete handoffs. A user or system may not notice the problem until validation, import, review, or execution. Even when the visible answer seems useful, the absence of required fields means the output does not satisfy the contract it was supposed to meet.
Structural Signal
A required field set exists, but the generated object does not fully instantiate it. The issue is not invalid syntax; it is an incomplete declaration relative to the schema, workflow, or consumer requirement.
Common Triggers
- Required fields are implied by examples but not stated as hard requirements
- The model treats empty, unknown, or unavailable values as permission to omit a field
- The output format is requested without a validation checklist
- Long schemas cause lower-salience fields to be dropped
- The prompt asks for concision and completeness at the same time
- The workflow does not declare placeholder behavior for unavailable required values
When to Use This Issue
Use this Issue when the output is missing fields that are required by the expected schema, template, form, parser, workflow, or downstream consumer.
When Not to Use This Issue
Do not use this Issue when fields are present but contain wrong values, wrong types, or hallucinated content. Do not use it when the entire output is malformed and cannot be parsed at all.