codex-cli

Automate Codex CLI code analysis, review, and validation tasks.

10|Updated Nov 21, 2025
One-click install
npx skills add https://github.com/creator-hian/claude-code-plugins --skill codex-cli-creator-hian
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-cli
Source: https://github.com/creator-hian/claude-code-plugins/tree/main/codex-plugin/skills/codex-cli
Command: npx skills add https://github.com/creator-hian/claude-code-plugins --skill codex-cli-creator-hian

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines code analysis, review, and validation using OpenAI Codex CLI, enabling efficient integration into CI/CD pipelines and automation scripts. It addresses the need for consistent code quality and security checks without manual intervention, saving developer time and improving reliability.

Core Features & Use Cases

  • Code Analysis & Review: Execute Codex commands for in-depth code examination, security audits, and architectural reviews.
  • Model & Reasoning Configuration: Select optimal Codex models (e.g., gpt-5.2 for complex tasks) and adjust reasoning effort for tailored analysis.
  • CI/CD Integration: Embed Codex into automation scripts and pipelines for continuous quality and security checks.
  • Use Case: Automatically review a pull request for security vulnerabilities and coding best practices before merging, generating a structured report for further processing.

Quick Start

Review this code for bugs: codex exec -s read-only "Review this code for bugs"

Frequently Asked Questions about codex-cli

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

FAQPage Schema
How do I automate code analysis in CI/CD pipelines?

Code analysis automation integrates Codex CLI into CI/CD workflows to perform continuous security reviews and quality checks without manual intervention. Use `codex exec` with sandbox modes (read-only, workspace-write) and configurable models to embed analysis into deployment pipelines, generating structured reports for automated decision-making.

Can I use Codex CLI for security vulnerability scanning?

Yes, Codex CLI performs automated security audits and vulnerability scanning across codebases. Run `codex exec -s read-only` to review code for security issues in Git and non-Git workspaces, supporting multi-directory analysis with `--add-dir` flags for comprehensive coverage.

What models and reasoning levels does Codex CLI support?

Codex CLI supports multiple models (gpt-5.2-codex, gpt-5.2, gpt-5.1-codex-max, gpt-5.1-codex-mini) with configurable reasoning effort levels. Select models based on task complexity—gpt-5.2 for complex analysis—and adjust reasoning intensity to balance analysis depth with execution time.

How do I run code analysis in non-TTY environments?

Codex CLI operates in both interactive terminals and non-TTY environments, making it suitable for automation scripts and headless systems. Use conventional input/output handling with sandbox modes and session management to execute analysis tasks reliably without terminal interaction.

Can I add multiple directories to a code review?

Yes, use the `--add-dir` or `-C` flags to include multiple directories in code analysis, enabling cross-module architecture assessments and security reviews. This supports multi-directory integration for comprehensive codebase evaluation in a single Codex CLI execution.

What sandbox modes protect my workspace during analysis?

Codex CLI offers three sandbox modes: read-only (safe inspection), workspace-write (controlled modifications), and danger-full-access (unrestricted). Choose read-only for security audits and reviews, or workspace-write for analysis tasks requiring code modifications within controlled boundaries.