displayName
OpenAI Structured Output (strict) guard
appliesTo
tool-result
phase
post-call
detectorKinds
onTrigger
block
severity
block-call
description
OpenAI Responses/Chat Completions support `response_format: { type:
"json_schema", json_schema: { ..., strict: true } }`. When strict=true
the server enforces JSON Schema constraints on the assistant output
and refuses to emit non-conforming results, surfacing a structural
error rather than a malformed response. Doc: OpenAI Structured Outputs.