Codex CLI Config
Reference for OpenAI Codex CLI configuration, permissions, sandboxing, and autonomy — sandbox_mode and permission profiles, approval_policy, Rules (.rules / prefix_rule), config.toml precedence and trust, AGENTS.md discovery, task design (Mini Codex, Done When, Plan mode), Skills/Plugins/subagents/MCP, and running Codex unattended in CI or Automations. Use when configuring Codex CLI, loosening or tightening its sandbox, diagnosing why a config.toml, AGENTS.md, or Rules file is not taking effect, or writing task prompts and AGENTS.md content Codex will act on well. Also applies when Codex CLI i
Install
/plugin marketplace add tomada1114/tomada-claude-pluginsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add tomada1114/tomada-claude-plugins --skill codex-cli-configgit clone --depth 1 https://github.com/tomada1114/tomada-claude-plugins.git /tmp/tomada-claude-plugins
mkdir -p ~/.claude/skills
cp -r /tmp/tomada-claude-plugins/tomada-kit/skills/codex-cli-config ~/.claude/skills/codex-cli-configPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Codex CLI configuration Reference knowledge base for OpenAI Codex CLI's configuration and permission system. All facts below and in `references/` were checked against the official docs (`learn.chatgpt.com/docs/*`) and the installed CLI's own `--help` output — prefer this over recalled knowledge, which is likely to carry the CLI's frequent renames and removals (see `references/diagnostics.md` for the version/date this was last verified against). ## Mental model Three axes, and they answer different questions: - **`sandbox_mode` (or the newer `[permissions.<name>]` profile system)** — *what* Codex can touch without asking: filesystem and network. Pick one of the two systems, never both — see `references/permissions-and-sandbox.md`. - **`approval_policy`** — *when* it has to ask before touching something outside that boundary — `references/permissions-and-sandbox.md`. - **Rules (`.rules` / `prefix_rule()`)** — narrow, per-command-prefix exceptions to the asking, layered user → project → admin, strictest always wins — `references/rules-execpolicy.md`. Config values themselves come from a layered `config.toml` (CLI flags > project `.codex/config.toml` > `--profile` file > user `~/.codex/config.toml` > system > built-in defaults) gated by per-project trust — `references/config-files-and-precedence.md`. ## Headline answers **"Let git commit run freely without `danger-full-access`."** `workspace-write`
Details
- Skill name
codex-cli-config- Repository
- tomada1114/tomada-claude-plugins
- Path
tomada-kit/skills/codex-cli-config/SKILL.md- Repo updated
- 12 November 2026
FAQ
What does the Codex CLI Config skill do?
Reference for OpenAI Codex CLI configuration, permissions, sandboxing, and autonomy — sandbox_mode and permission profiles, approval_policy, Rules (.rules / prefix_rule), config.toml precedence and trust, AGENTS.md discovery, task design (Mini Codex, Done When, Plan mode), Skills/Plugins/subagents/MCP, and running Codex unattended in CI or Automations. Use when configuring Codex CLI, loosening or tightening its sandbox, diagnosing why a config.toml, AGENTS.md, or Rules file is not taking effect, or writing task prompts and AGENTS.md content Codex will act on well. Also applies when Codex CLI i
How do I install the Codex CLI Config skill?
Claude Code plugin: /plugin marketplace add tomada1114/tomada-claude-plugins. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Codex CLI Config popular?
It lives in tomada1114/tomada-claude-plugins, which has 5 GitHub stars and 0 forks.
Related skills
Auditing Prompt Docs
5tomada1114/tomada-claude-plugins
Audit and fix prompt-bearing documents — SKILL.md, CLAUDE.md, AGENTS.md, agent and command definitions, reference files, whole repositories — against Anthropic's published prompting guidance for current Claude models, proposing edits for approval before writing anything. Also answers how to prompt a specific model: holds the per- model guidance for Claude Fable 5.2, Claude Opus 5.5, Claude Sonnet 5.5, and Claude Haiku 5.5 alongside the cross-model practices. Use when reviewing or slimming an instruction file, when instructions are ignored or fire too often, when migrating a prompt to a newer m
Auditing Repositories
5tomada1114/tomada-claude-plugins
Audit the repository you are in, often a template or an app cut from one, for technical debt and room to improve in four areas: architecture and design patterns; tests; file structure and code quality; and tooling, CI, dependencies and the agent harness. Design a safe, test-first fix for every evidenced finding, and file the findings as GitHub issues that follow the repository's own issue conventions (triaging-issues, issue forms, past issues). A finding inherited from the template is also filed in the template repository. Cheap read-only sub-agents collect the evidence and high-effort analyst
Authoring Goal Prompts
5tomada1114/tomada-claude-plugins
Author a self-contained prompt for Claude Code's /goal command, run unattended in a separate session — a lead-to-staff handoff where this session researches, designs, and decides with full context, and a fresh executor session (Opus 5.5 or Fable 5.2, with none of that context) implements. Drafts a goal with measurable done-criteria, transcript-verifiable checks, scope + anti-cheat constraints, and stop rules; packages design intent, code examples, research findings, checklists, and pre-answered decisions as support files in a per-goal state directory whenever the executor would otherwise re-de
Commit
5Magnus-Gille/claude-skills
Create or amend a Git commit with author verification, diff review, and a conventional message. Use only when the user explicitly invokes $commit or asks to commit or amend changes.
Code Friction
5Magnus-Gille/claude-skills
Capture a small, opt-in set of concrete code-changeability observations during a coding task and hand them to Close without affecting readiness.
Designing Architecture
5tomada1114/tomada-claude-plugins
Decide a new app's hard-to-reverse design choices from its signed-off requirements and UX drafts before any issue is filed — design lock, app shape, sandbox and OS permissions, persistence, backend and sync, auth, key dependencies, locales, distribution — through researched option rounds with the user, and write each decision in the present tense, with its reasoning, into the repository skill that owns the subject (its SKILL.md or references/), or AGENTS.md's Architecture section for a layer rule. Creates no architecture overview, roadmap, decision record, or ADR; a subject with no owning skil
Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.