codex

Codex Critique

Run OpenAI's Codex CLI to get an independent critique of a file (spec, code, etc.).

Pre-flight

Verify codex is available: !which codex

If codex is not found, tell the user to install and authenticate (codex login).

Usage

/critique:codex $ARGUMENTS

• $0 — optional file path to critique (relative or absolute)
• $1 — optional focus area (e.g., "security", "performance", "UX gaps")
• --model <model> — override the codex model for this invocation

All arguments are optional. When no file is given, infer the target from conversation context.

Configuration

Config is resolved with the following precedence (first match wins):

1. CLI flag (--model) — one-off override
2. Local config (.claude/skill-configs/codex/config.local.yaml) — personal/local scope, gitignored
3. Project config (.claude/skill-configs/codex/config.yaml) — project scope, committed to repo
4. Default — gpt-5.4

model: gpt-5.4  # model to use with codex exec

See config.example.yaml in the critique plugin's codex skill for reference.

Setup

1. Check if $ARGUMENTS contains --model <model>. If so, use that model and skip config lookup.
2. Check for config files (first match wins):
   • .claude/skill-configs/codex/config.local.yaml (local scope, gitignored)
   • .claude/skill-configs/codex/config.yaml (project scope, committed to repo)
3. If no config found, use gpt-5.4 as the default.
4. Set ${MODEL} to the resolved model name.

Reasoning Effort

The user can request higher reasoning effort by saying things like "xhigh", "high effort", "think harder", or "deep review". Map these to the --config model_reasoning_effort flag:

User says	Flag value
(default)	(omit flag — uses codex default "medium")
"high"	high
"xhigh", "maximum", "think hard"	xhigh

How It Works

1. Determine what to critique — use one of these strategies, in priority order:
   • If $0 is provided, use it as the file path
   • If the user mentions a file by name or path (e.g., @games/foo/bar.js), use that
   • If changes were made during the current conversation (specs written, code edited, etc.), critique those — use git diff against the state before the conversation's changes to identify the affected files and pass them to codex
   • If the user says something like "critique this" or "get a second opinion", look at the most recent substantive work product in the conversation
   • If nothing can be inferred, ask the user what to critique
2. Resolve the working directory (project root)
3. Build a review prompt tailored to the content type — use the templates in ${CLAUDE_SKILL_DIR}/templates/:
   • spec-review.md — for specification files
   • code-review.md — for source code files
   • diff-review.md — for reviewing recent changes
4. Run codex exec non-interactively with read-only sandbox
5. Present the findings to the user

Invocation

codex exec \
  -m ${MODEL} \
  -s read-only \
  -C <project-root> \
  --config model_reasoning_effort="<effort>" \
  "<review-prompt>"

Flags:

• -m ${MODEL} — the resolved model (default: gpt-5.4)
• -s read-only — read-only sandbox (no file modifications)
• -C <dir> — set working directory so codex can read referenced files
• --config model_reasoning_effort="<effort>" — reasoning depth: medium (default), high, or xhigh. Omit for default. Higher effort = slower but more thorough analysis.

Timeout: 300 seconds at default effort, 600 seconds at high/xhigh (codex does more reasoning passes)

After Running

1. Read the full codex output (it may be persisted to a temp file if large)
2. Present findings to the user in a concise summary
3. If the user wants to act on the feedback, proceed with edits as a separate step

Notes

• If codex errors, check codex --version and suggest codex login if auth fails