codex-review

Codex Review

Use Codex CLI to get a second opinion. All execution goes through review.js with --sandbox read-only.

Locating review.js

Find the script with:

printf '%s\n' ~/.claude/plugins/cache/second-opinion-skill/second-opinion-skill/*/bin/review.js 2>/dev/null | sort -V | tail -1

Store the result as REVIEW_SCRIPT. Do not call codex directly.

Model selection (optional)

Codex uses its configured default model if no model is specified. Ask the user whether they want to specify a model or use the default. If they choose to specify, prompt for the model name (they must know it; there is no listing command).

Pass the model inline as --engine=codex:<model> if provided. Use bare --engine=codex for the default.

Determining what to review

Pass the appropriate flag to review.js (it handles fetching and embeds diff/file content inline before the prompt):

What to review	Flag
Unstaged changes	--diff=unstaged
Staged changes	--diff=staged
Last commit	--diff=last-commit
Branch vs main	--diff=branch
Custom range	--diff="HEAD~3..HEAD"
Specific file	--file=<absolute-path>
General question	(no flag)

Running

Without model (use default):

"$REVIEW_SCRIPT" --engine=codex --cwd=<repo-path> [--diff=<spec>|--file=<path>] "<review template>"

With a specific model:

"$REVIEW_SCRIPT" --engine=codex:<model> --cwd=<repo-path> [--diff=<spec>|--file=<path>] "<review template>"

Composing the prompt

For the full guidance on how to compose the prompt — when to embed the user's ongoing task / context (Tier A), when to fall back to default templates (Tier B), and when to use --no-embed for very large diffs (Tier C) — read the second-opinion skill. The default templates live there too.

Safety toggle

By default review.js applies this engine's read-only / sandbox / plan-mode flags. Pass --unrestricted only when the engine genuinely needs to edit files or run commands; review.js will drop the safety flags and log a stderr warning.

Output envelope

review.js wraps every prompt with <<<SECOND_OPINION_START>>> / <<<SECOND_OPINION_END>>> markers and asks the engine to emit its real answer between them. After reading the log file, extract the text between the markers — that is the clean payload, free of reasoning traces and tool noise. Pass --no-wrap to disable.

Capturing output

When review.js runs from an agent harness (non-TTY stdout), engine output is not streamed to stdout. It's written to a temp file, and stdout receives only a banner pointing at the log path:

LOG FILE: /var/folders/.../second-opinion-codex-<ts>.log

After the command exits, use the Read tool on that path to get the full Codex review. Don't pipe to | tail -N or | head -N — the engine output isn't on stdout in this mode. Pass --log=<path> for a known location; pass --log=- to restore tee-to-stdout behavior.

Presenting results

Show Codex's full response under a ## Codex's Take heading (include model name if one was specified: ## Codex's Take (<model>)). Don't filter or summarize. If issues are raised that need fixing, address them and note what changed.