feat(api): add prompt-cache prewarming

[jbeckwith-oai]

Summary

Adds the optional prompt-cache prewarming flag to stable and beta response creation commands.

Changes

• Support --prompt-cache-options.prewarm and the equivalent object and YAML options
• Preserve omitted options and explicitly supplied false and true values

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-17T19:04:37.114589Z	8b69add	PR opened
🔒 Security Review	✅ Completed 2026-09-17T19:05:21.618670Z	8b69add	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[github-actions]

Castiron custom code

✅ No new custom-code files detected.

5 mixed files remain; 0 existing customizations changed.

Compared 20db45ee515e → 8b69add54b04. Generated baselines verified.

5 existing customizations unchanged

• pkg/cmd/adminorganizationcertificate.go
• pkg/cmd/cmd.go
• scripts/castiron/README.md
• scripts/castiron/custom_code_report.py
• scripts/castiron/test_custom_code_report.py

A changed generated baseline means this report cannot reliably identify which handwritten lines changed.

Inspect the custom-code diff

Download the exact patch produced by this run (requires repository access):

gh run download 35262524698 --repo openai/openai-cli \
  --name castiron-custom-code-35262524698-1 --dir /tmp/castiron-custom-code-35262524698-1
git apply --stat /tmp/castiron-custom-code-35262524698-1/custom-code.patch
cat /tmp/castiron-custom-code-35262524698-1/custom-code.patch

Or reproduce it from an SDK checkout containing the vendored reporter:

git fetch --no-tags origin 20db45ee515e2f7f1d47f634a1b5699e20b548ac 8b69add54b04dd6bc07f6841eb57cf55d141931f
python3 scripts/castiron/custom_code_report.py report \
  --base 20db45ee515e2f7f1d47f634a1b5699e20b548ac \
  --head 8b69add54b04dd6bc07f6841eb57cf55d141931f --fetch --require-head-hash --public \
  --out /tmp/castiron-custom-code-8b69add54b04
cat /tmp/castiron-custom-code-8b69add54b04/custom-code.patch

This is the current full custom patch for mixed files, not an attribution of only the handwritten lines changed by this PR.

Full report and patch

[markstuart-oai]

Reviewed 8b69add54b04dd6bc07f6841eb57cf55d141931f, including an independent CLI review and the complete supporting schema refresh. No actionable findings. Stable and beta creation both register the optional boolean through the existing inner-flag machinery; the schema definitions agree with the canonical optional field. The source trace preserves omission, explicit false/true, sibling cache options and caller-supplied request data. Server-side prewarm behavior is not reimplemented in the CLI, and no new abstraction or shared parsing change is introduced.

CI run 35262485829 passed build/artifact checks, lint, the CLI/internal suite against Steady, and Windows checks. Its tested merge 68fef721b8c67c5dacf98ead39cc17338ef418bf has the same tree as this head. The changed fixtures exercise true through object, dotted-flag and YAML input; formatting also passed locally. Local vet stopped before analysis because the pinned dependencies were unavailable, so no independent local loopback matrix or live API pass is claimed.

Approved.

[saioai]

the optional flag works in stable and beta, preserves false and omitted values, and keeps the other request options intact. independent checks and CI pass.