Skip to content
Back to skills

Codex Cli 1

ASecurity

Use Codex CLI for web search, code review, and writing review. Triggers on (1) web search requests, (2) code review of git changes, (3) writing/file review. Use for "search the web for...", "review this code", "review my changes", "review this file".

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 27, 2026
toolsbashgitsecurity

Works with

  • cli

Security analysis

A100/100

Scanned September 27, 2026

npx -y skills add David-Li0406/meta-skill-evloving --skill codex-cli-1 --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Codex Cli 1?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Codex Cli 1
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/david-li0406-codex-cli-1/badge)](https://www.skillsdirectory.com/skills/david-li0406-codex-cli-1)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: codex-cli
description: Use Codex CLI for web search, code review, and writing review. Triggers on (1) web search requests, (2) code review of git changes, (3) writing/file review. Use for "search the web for...", "review this code", "review my changes", "review this file".
---

# Codex CLI

OpenAI's Codex CLI for web search and reviews. All commands are non-interactive.

## Web Search

ChatGPT-tuned agentic search. Default: `gpt-5.2` with medium reasoning.

```bash
codex exec "your question" -c 'tools.web_search=true' -m gpt-5.2 -c model_reasoning_effort="medium" \
  --output-last-message /tmp/codex-result.txt >/dev/null 2>&1 && cat /tmp/codex-result.txt
```

## Code Review (Git Changes)

Default: `gpt-5.2-codex` with high reasoning.

```bash
# Review uncommitted changes (staged + unstaged + untracked)
codex review --uncommitted -c model="gpt-5.2-codex" -c model_reasoning_effort="high"

# Review against a base branch
codex review --base main -c model="gpt-5.2-codex" -c model_reasoning_effort="high"

# Review a specific commit
codex review --commit <sha> -c model="gpt-5.2-codex" -c model_reasoning_effort="high"

# Custom review instructions
codex review --uncommitted "Focus on security issues" -c model="gpt-5.2-codex" -c model_reasoning_effort="high"
```

## Writing/File Review

Review a file for clarity, density, and fluff removal.

```bash
codex exec "Review the writing at <filepath>. Provide an improved version in full, then explain each change. Prioritize: removing fluff, increasing information density, improving clarity. No stylistic preferences." \
  --skip-git-repo-check --output-last-message /tmp/codex-result.txt >/dev/null 2>&1 && cat /tmp/codex-result.txt
```

Use your judgment to incorporate high-value changes directly. Briefly summarize what you changed and why.

## Reasoning Effort

Override with `-c model_reasoning_effort="low|medium|high"`. Higher = deeper analysis, slower.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…