ci(codex-review): bump reviewer model → gpt-5.6-sol (LAB-5088)

[nsportsman]

Part of the fleet-wide reviewer-model rollout (LAB-5088): bumps the codex-review job's model to gpt-5.6-sol. Validated by palatine#324 (ran gpt-5.6-sol on the identical codex runtime 0.135.0) and public-workflows#145 (default bumped). This is a one-line change; no reusable-workflow SHA bump is needed since all reusable versions pin the same codex-version. Merge is a human decision.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 10 minutes

Your organization has reached its usage spending cap. Adjust your spending cap in the billing tab.

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: d81987cf-539c-40d9-84ff-7af02e98afc3

📥 Commits

Reviewing files that changed from the base of the PR and between 8e262bf and 07027f7.

📒 Files selected for processing (1)

• .github/workflows/codex-code.yml

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch lab-5088-codex-sol

---

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

Gemini Review

Critical Issues

None.

Security

No security concerns flagged.

Suggestions

No suggestions.

No critical issues — LGTM pending human review.

---

Reviewed by Gemini (gemini-3.1-pro-preview)

[github-actions]

Codex Review

Critical Issues

• .github/workflows/codex-code.yml:26 selects gpt-5.6-sol, but the pinned reusable workflow runs Codex CLI 0.135.0; OpenAI requires CLI 0.144.0 or newer for GPT-5.6. Reviews may fail instead of posting feedback. Upgrade the centralized workflow/CLI first or retain gpt-5.5. (help.openai.com)

Security

No security concerns flagged.

Suggestions

No suggestions.

---

Reviewed by Codex (gpt-5.6-sol)

[github-actions]

Claude Review

Critical issues

None. Single-line change swapping the model input from gpt-5.5 to gpt-5.6-sol. The reusable workflow stays pinned to SHA 3fb9ffb (v2.11.3), so no security posture or contract shifts here. One thing worth a human sanity-check: confirm gpt-5.6-sol is a valid/accepted value for the reusable workflow's model input at that pinned SHA — an unknown model string would only surface at runtime, not at config-parse time.

Security

No security concerns flagged. Secret handling (OPENAI_API_KEY), permissions, and the pinned reusable-workflow SHA are all unchanged.

Test coverage

N/A — CI workflow config only, no production code changed.

[nsportsman]

Automated-review adjudication (LAB-5088) — one-line model: bump in the codex-review caller (.github/workflows/codex-code.yml, gpt-5.5 → gpt-5.6-sol). Adjudicated per adjudicating-review-findings.

Codex — "CRITICAL: gpt-5.6-sol requires Codex CLI ≥ 0.144.0; the reusable runs 0.135.0, reviews may fail instead of posting feedback" → REFUTED.

• Self-refuting. This review is footered "Reviewed by Codex (gpt-5.6-sol)" — produced by gpt-5.6-sol on the pinned Codex CLI 0.135.0 (codex-action@c25d10f3). The combination it calls unsupported is the one that produced the comment.
• Sibling PRs, identical runtime. janus-framework#29, nero#41, pius#103 each posted "No critical issues — Reviewed by Codex (gpt-5.6-sol)" on the same 0.135.0 pin.
• Canary. palatine#324 set gpt-5.6-sol; codex-review PASSED, tagged Reviewed by Codex (gpt-5.6-sol) on codex-version 0.135.0 / codex-action@c25d10f3.
• Access. All three GPT-5.6 tiers (luna/terra/sol) validated accessible against the reviewer's own OPENAI_API_KEY on 2026-07-22; public-workflows#145 (merged) set gpt-5.6-sol as the shared default. GPT-5.6 shipped 2026-06-23 as luna/terra/sol — no plain gpt-5.6.
• The cited help.openai.com article ("GPT-5.6 in ChatGPT") is about the ChatGPT product and sets no Codex-CLI version floor; the "0.144.0" requirement is unsupported by it.
• codex-review is green on this PR.

Claude — "confirm gpt-5.6-sol is a valid/accepted model input; an unknown string would surface only at runtime, not config-parse time" → REFUTED (advisory sanity-check).

• The identifier is validated by the evidence above and already executed on this PR's own codex-review run. "Pass-through without validation/fallback" is pre-existing behavior of the caller workflow, not introduced by this one-line diff → out of scope for this PR.

Gemini → LGTM. CodeRabbit "review rate limited" = throttle, not a diff finding.

No confirmed findings → no code change, nothing deferred. CI green. Stopping for human merge review (not merging).