Codex CLI Skill

Trust Score 80/100

Operational guide for running the Codex CLI: sandbox modes, flags, resume workflows, and safe defaults for model selection and stderr handling.

triggers:codex execcodex resumesandbox--skip-git-repo-check

GitHub SKILL.md

What it does

This skill documents best-practice usage of the Codex CLI for code analysis and automated editing tasks. It prescribes sandbox modes (read-only, workspace-write, danger-full-access), how to assemble commands safely (--skip-git-repo-check, piping resumes), and stderr handling to suppress thinking tokens. The guide is practical and procedural, focused on safe execution rather than modelling.

When to use it

Use this skill when the agent needs to invoke the Codex CLI to analyze or modify code, run resume sessions, or perform sandboxed operations. It's especially relevant when you must avoid specifying explicit models and when you need to suppress or surface stderr tokens intentionally.

What's included

Scripts: none bundled, the SKILL.md provides command examples and flags.
References: CLI usage patterns and resume examples are included in the body.
Instructions: Step-by-step rules for choosing sandbox mode, building codex exec commands, resuming sessions, and error-handling guidance.

Compatible agents

Relevant to agents that orchestrate code CLIs (Codex/Copilot/Codex-backed workflows) and CI automation where controlled sandboxing is required.

Audit Summary

Codex CLI operational guide covering sandbox modes, flags, resume workflows, and safe defaults. No bundled scripts — purely instructional. Clear quick reference table and step-by-step instructions, but limited to OpenAI Codex CLI users with a ChatGPT account. Suppresses stderr by default which could mask errors.

Watch Out

Only applicable when using Codex CLI with a ChatGPT account — explicit model flags fail
2>/dev/null hides stderr including errors — can mask real issues
Requires AskUserQuestion tool which may not be available in all agent contexts

Notes

Clean, instructional-only skill with no executable code. Niche but well-documented for its target audience. The skip-git-repo-check flag is mentioned as always-use which slightly lowers security posture but is documented as a deliberate choice.

Information

Repository: claude-skills
Stars: 180
Installs: 0

Trust Score

Overall80

Security95

Code Quality62

Architecture70

Usefulness65

Related Skills

Development Worktree

Create an isolated git worktree for feature work, auto-run project setup, and verify a clean test baseline before development.

Readwise Reader Document Management

Manage Readwise Reader documents: list, save, search, move, tag, highlight, export and bulk-edit via official and custom CLIs.

Bounty Hunter — Atlas

Persona skill: 'Atlas' — a profit-focused developer persona for discovering, evaluating and executing paid bounties or freelance tasks with ROI-aware workflows.

Junshi — Research Advisor

Daily strategic research advisor that scans arXiv/venues, digests papers, and proposes bold, ranked research ideas tailored to the user's profile.

Full Stack Builder

End-to-end builder that scaffolds, implements, tests, and optionally deploys web and API applications from a natural-language specification.

ezBookkeeping API Tools

Command-line API tools for ezBookkeeping: record and query transactions, retrieve accounts/categories/tags, and fetch exchange rates for self-hosted personal fi

Feishu Voice Sender

Convert MP3s and send them as native Feishu voice messages (playable voice clips) to users or groups.

Claw Bench

Benchmarking skill that guides an agent through a structured suite of capability tests and reporting steps for leaderboard submission.

Back to Skills