Claude Code skill to delegate prompts to OpenAI Codex CLI for code analysis, refactoring, code review, and automated editing
Enable Claude Code to invoke the Codex CLI (codex exec and session resumes) for automated code analysis, refactoring, code review, and editing workflows.
Fork of skills-directory/skill-codex with enhancements: updated models (gpt-5.4 default), new flags (--add-dir, --ephemeral, -i images, codex exec review), and printf for safe prompt piping.
codex CLI installed and available on PATH.codex --version; resolve any errors before using the skill.This repository is structured as a Claude Code Plugin with a marketplace. You can install it as a plugin (recommended) or extract it as a standalone skill.
Install via Claude Code's plugin system for automatic updates:
/plugin marketplace add integralmedia1/skill-codex
/plugin install skill-codex@skill-codex
Extract the skill folder manually:
git clone --depth 1 [email protected]:integralmedia1/skill-codex.git /tmp/skill-codex-temp && \
mkdir -p ~/.claude/skills && \
cp -r /tmp/skill-codex-temp/plugins/skill-codex/skills/codex ~/.claude/skills/codex && \
rm -rf /tmp/skill-codex-temp
By default, this skill suppresses thinking tokens (stderr output) using 2>/dev/null to avoid bloating Claude Code's context window. If you want to see the thinking tokens for debugging or insight into Codex's reasoning process, explicitly ask Claude to show them.
User prompt:
Use codex to analyze this repository and suggest improvements.
Claude Code response: Claude will activate the Codex skill and:
gpt-5.4, gpt-5.3-codex-spark, gpt-5.3-codex, or gpt-5.2) unless already specified in your prompt.low, medium, high, or xhigh) unless already specified.read-only for analysis).codex exec -m gpt-5.4 \
--config model_reasoning_effort="high" \
--sandbox read-only \
--skip-git-repo-check \
"Analyze this repository comprehensively..." 2>/dev/null
Result: Claude will summarize the Codex analysis output, highlighting key suggestions and asking if you'd like to continue with follow-up actions.
| Model | Best For |
|---|---|
gpt-5.4 | Default, strongest reasoning |
gpt-5.3-codex-spark | Fast agentic coding |
gpt-5.3-codex | Balanced agentic coding |
gpt-5.2 | General frontier model |
| Level | Best For |
|---|---|
high | Default, deep analysis and quality output |
xhigh | Maximum quality, longest thinking time |
medium | Quick daily tasks |
low | Fast, simple operations |
| Mode | Behavior |
|---|---|
read-only | No file modifications (default for analysis) |
workspace-write | Can edit files in workspace |
danger-full-access | Full network and filesystem access |
Codex sessions persist and can be resumed:
# Resume most recent session
printf '%s' "continue with..." | codex exec --skip-git-repo-check resume --last
# Resume interactively
codex resume --last
# Fork a previous session
codex fork --last
See plugins/skill-codex/skills/codex/SKILL.md for complete operational instructions, CLI options, and workflow guidance.
MIT License - see LICENSE for details.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub integralmedia1/skill-codex --plugin skill-codexIntegral Media brand style guide for HTML documents, reports, presentations, and Google Sheets formatting
integralmedia1Claude Code skill to delegate prompts to Google Gemini CLI for code analysis, refactoring, codebase investigation, and multi-model collaboration
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
nyldnPersistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
OthmanAdiClaude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.
Chachamaru127Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
affaan-mPlugin-safe Claude Code distribution of Antigravity Awesome Skills with 1,561 supported skills.
sickn33