01 / DECIDE
Architecture decision
Use it for draft illustrations, product variations, storyboards, and controlled creative tools with a human acceptance step.
Separate prompt intent, generation request, moderation state, asset storage, metadata, and editorial acceptance into observable stages. Map input, output, state, and side effects as one observable path before deciding which layer owns put image generation behind an asset pipeline.
02 / BUILD
Three-step implementation
- 01
Frame the contract
Write down the caller, data classification, success condition, timeout, cancellation, and ownership. Use it for draft illustrations, product variations, storyboards, and controlled creative tools with a human acceptance step.
- 02
Build one narrow path
Implement one end-to-end path with request correlation, typed state, and reversible failure handling. Separate prompt intent, generation request, moderation state, asset storage, metadata, and editorial acceptance into observable stages.
- 03
Prove the outcome
Turn acceptance into a repeatable fixture, contract test, or browser test. Reviewers can trace each accepted asset to prompt version, model configuration, moderation result, and final transformation.
03 / BOUND
Production boundary
Track rights, consent, provenance, and retention; never imply generated media is a verified photograph or historical record.
04 / PROVE
Acceptance evidence
Reviewers can trace each accepted asset to prompt version, model configuration, moderation result, and final transformation.
SOURCE / HTTP
Reproducible source probe
curl -fsSI 'https://developers.openai.com/api/docs/guides/image-generation' | sed -n '1,5p'