メニュー
Inverted orchestration—Codex CLI acts as CTO (planning and reviewing), Claude Code executes with native tools. Use when a task benefits from external architectural oversight before and after implementation.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/tdimino/claude-code-minoan --skill codex-ctoコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Search, post, monitor, and archive on Twitter/X via five tools: the official hosted X MCP server (full-archive search, trends, bookmarks, Articles via xurl bridge), x-search CLI (cost-tracked API v2 research with calendar-day windows, volume counts, multi-query fan-out, feeds, watchlists, posting), xurl (official CLI for any endpoint incl. media/DMs), bird CLI (free session-based reads/writes, frozen at 0.8.0), and Smaug bookmark archival. Triggers on: search tweets, tweets from today, what did X tweet, latest tweets from, tweet volume, expand a Twitter query, check Twitter for, post a tweet, tweet this, post to X, reply on Twitter, full-archive search, what's trending, check mentions, my mentions, timeline, DMs, monitor account, research topic on Twitter, my bookmarks, archive bookmarks.
Manage Claude Code sessions — search by topic or ID, browse recent sessions with full metadata (tags, summaries, titles, cost), view title/nickname history timelines, resume in Ghostty tabs, spawn interactive or headless sessions, monitor live sessions, and bootstrap new setups. Triggers on resume session, find session, list sessions, recent sessions, spawn session, session history, what was I working on, open in ghostty, title history, session nicknames.
Summon Claude Fable 5 (Mythos-class, xenos daimon) for long-horizon tasks. Handles model override, cost awareness, and task routing. Availability probed at invocation.
SKILL.md を表示中
SOC 職業分類に基づく
| name | codex-cto |
| description | Inverted orchestration—Codex CLI acts as CTO (planning and reviewing), Claude Code executes with native tools. Use when a task benefits from external architectural oversight before and after implementation. |
| argument-hint | ["objective"] |
| disable-model-invocation | true |
Codex plans and reviews. Claude Code executes. This inverts the codex-orchestrator pattern: instead of Claude Code delegating to Codex subagents, Codex directs Claude Code as the engineering team.
1. CTO plans → cto-invoke.sh plan "<objective>"
2. You execute → native tools (Read, Edit, Write, Bash, Glob, Grep)
3. CTO reviews → cto-invoke.sh review "<objective + results>"
4. Loop → if "revise", fix and re-submit for review (max 5 iterations)
Run the plan phase with the user's objective:
~/.claude/skills/codex-cto/scripts/cto-invoke.sh plan "<OBJECTIVE>" [--model MODEL]
Parse the returned JSON. It contains:
objective — what to accomplishanalysis — CTO's assessment of the codebasestrategy — high-level approachtasks[] — ordered task list with IDs, actions, targets, descriptions, acceptance criteria, and dependenciesverification — commands to run after all tasks completeWork through each task in dependency order using native tools. For each task:
description and acceptance_criteriacreate_file, modify_file, delete_file, run_command, verify)Do not deviate from the plan without reason. If a task is blocked or the plan has an error, note it and continue with unblocked tasks. Report blockers in the review submission.
After executing all tasks, run the review phase. Include the original objective and a summary of what was done:
~/.claude/skills/codex-cto/scripts/cto-invoke.sh review "<OBJECTIVE>. Results: <summary of what was done, any issues encountered>"
Parse the returned JSON:
verdict: "approve" — done. Report success to the user.verdict: "revise" — check task_reviews for feedback and revised_tasks for new/updated tasks. Fix issues and re-submit.verdict: "abort" — fundamental problem. Report the CTO's reason to the user and stop.On a "revise" verdict:
task_reviews entry with status fail or partialfeedback and any revised_tasksThe --model flag passes through to codex exec. Available models:
| Model | ID | Notes |
|---|---|---|
| GPT-5.4 Pro | gpt-5.4-pro | Default for plan phase. Deepest reasoning for architectural decomposition. |
| GPT-5.4 | gpt-5.4 | Default for review phase. Unified coding + reasoning flagship. |
| GPT-5 Mini | gpt-5-mini | Cost-optimized, fast iteration. |
| GPT-5.3-Codex | gpt-5.3-codex | Previous generation. Still functional. |
Omit --model to use the default from ~/.codex/config.toml.
| Option | Description |
|---|---|
--model <id> | Override model (default: config default) |
--dry-run | Print command without executing |
Plan and review JSON files are saved to .codex-cto/<PID>/ in the working directory. Each invocation creates sequentially numbered files (plan-001.json, review-001.json, etc.). PID namespacing prevents conflicts between concurrent sessions.
Reuse codex-orchestrator's status script for CLI diagnostics:
~/.claude/skills/codex-orchestrator/scripts/codex-status.sh
# Simple feature
/codex-cto Add a health check endpoint at /health returning {status: "ok"}
# With model override
/codex-cto Refactor the auth module to use JWT --model gpt-5-mini
# Dry run to inspect the command
~/.claude/skills/codex-cto/scripts/cto-invoke.sh plan "Add rate limiting" --dry-run
This skill is the inverse of codex-orchestrator. Use codex-orchestrator when Claude Code should delegate one-shot specialist tasks to Codex. Use codex-cto when Codex should direct Claude Code through a plan-execute-review loop.