skills.rest
scsm-unrestrict
scsm-unrestrictCOMMUNITY
Explore all skills by this author →
GitHub

codex-review

Runs independent code reviews via the OpenAI Codex CLI and saves severity-prioritised reports.

1|Updated Aug 18, 2026
Search Tags:#git#code review#pull request#security audit#codex cli#second opinion#severity report
One-click install
npx skills add https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent --skill codex-review-scsm-unrestrict
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent/tree/main/frontend-engineer/skills/codex-review
Command: npx skills add https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent --skill codex-review-scsm-unrestrict

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Code written or reviewed by a single AI model carries author bias, so bugs, security holes, and regressions slip through. This Skill gets a second opinion from a different model family by running the OpenAI Codex CLI in headless review mode against your changes, a branch, a commit, or the whole app. ## Core Features & Use Cases - Scoped reviews: Review uncommitted changes, a branch diff against a base, a specific commit, or the entire codebase at HEAD using the appropriate codex review flag. - Structured severity reports: Saves a timestamped Markdown report to .jez/reviews/ with findings prioritised as critical, high, medium, low, and nit, each with file:line references. - Advisory workflow: Surfaces top findings inline and offers to fix specific issues only with user approval, never auto-applying suggestions. - Use Case: Before opening a PR, ask for an independent review of your branch versus main; the Skill runs Codex, saves the report, and shows you the critical and high findings with exact file locations. ## Quick Start Ask the agent to run a codex review of my uncommitted changes and show me the critical findings.

Frequently Asked Questions about codex-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a second opinion code review from a different AI model?▼

Run the OpenAI Codex CLI in headless review mode against your code. The review covers bugs, security issues, regressions, and testing gaps, and saves a severity-prioritised report with file:line references to .jez/reviews/.

How do I review uncommitted changes with the Codex CLI?▼

Pipe the review prompt into codex review with the --uncommitted flag. For a branch comparison use --base main, for a specific commit use --commit followed by the hash, and with no flag it reviews the whole app at HEAD.

What do I need installed to run codex review?▼

You need the OpenAI Codex CLI installed and authenticated. Check with 'which codex', install it via brew install codex on macOS or from the GitHub repository, and run codex login if the first review call fails with an auth error.

Does codex review automatically fix the issues it finds?▼

No, the review is advisory only. Findings are reported with severity levels and file:line references, and fixes are applied only after you discuss them and give explicit approval.

When should I not use an independent Codex code review?▼

Skip it for trivial changes like typos or one-line fixes, and for research or architecture discussions where a multi-model opinion tool fits better. It is designed for reviewing concrete code changes, not open-ended questions.

Related Skills

Explore More Software Engineering Skills→
rickyanandarickyananda
1

codex

Delegate coding tasks to OpenAI Codex CLI within git repositories.

Community
Advanced
bitwardenbitwarden
129

reviewing-incremental-changes

Re-review only changed lines in pull requests after developer updates.

Official
Basic
EremesNGEremesNG
1

verification

Run automated checks and manual review to verify completed work.

Community
Intermediate
Jesse-L-MJesse-L-M

codex

Generates code reviews, challenges, and consultations via the Codex CLI.

Community
Advanced