plan-and-build

plan-and-build is a skill for Claude Code from newrise0410/claude-codex-workflow. It costs 114 tokens per session (2,976 once invoked), scanned A, original, MIT.

A workflow that assigns planning, coding, and review to different AI models in sequence.

In plain words
What is it for?
Use it for larger feature work that needs a written plan, code changes, and a one-time review; it is not intended for tiny edits or open-ended debugging.
Why use it?
It separates repository research and planning from implementation, then adds an independent check of the result.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: names the AskUserQuestion tool; mentions Codex.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the claude-codex plugin — 1 skill, 2 agents shipped together

Good fit Use it for larger feature work that needs a written plan, code changes, and a one-time review; it is not intended for tiny edits or open-ended debugging.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add newrise0410/claude-codex-workflow
Claude Code
/plugin install claude-codex

Made for: Claude Code.

Or install claude-codex, the plugin that ships this one along with the rest of its 1 skill, 2 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for plan-and-build

README.md
[![agentmods](https://agentmods.dev/badge/skills/newrise0410/claude-codex-workflow/plan-and-build/github.svg)](https://agentmods.dev/skills/newrise0410/claude-codex-workflow/plan-and-build)
Your own site
<a href="https://agentmods.dev/skills/newrise0410/claude-codex-workflow/plan-and-build"><img src="https://agentmods.dev/badge/skills/newrise0410/claude-codex-workflow/plan-and-build/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for plan-and-build

Your own site · 80×15
<a href="https://agentmods.dev/skills/newrise0410/claude-codex-workflow/plan-and-build"><img src="https://agentmods.dev/badge/skills/newrise0410/claude-codex-workflow/plan-and-build.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 114 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,976 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00114 $0.02976
Opus 5 $0.00057 $0.01488
Sonnet 5 $0.00023 $0.00595
Haiku 4.5 $0.00011 $0.00298

Measured 9d ago against content hash bfe0979f9266, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

plan-and-build scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/plan-and-build/SKILL.md · 166 lines

How it starts

The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.

plan-and-build

기획·구현·검수를 서로 다른 모델에 나눠 맡긴다. 너(오케스트레이터)는 코드를 직접 쓰지 않고 단계를 진행시키고 결과를 사람에게 보고한다.

단계 담당 실행 방식 산출물
1. 기획 Claude Fable fable-planner 서브에이전트 PLAN.md
2. 구현 codex-luna-max codex-build 스크립트 BUILD_SUMMARY.md, DIFF.patch
3. 검수 Claude Fable fable-reviewer 서브에이전트 REVIEW.md

검수는 1회다. 자동 재작업 루프는 돌리지 않는다. FAIL이 나오면 사람에게 보고하고, 수정 반영은 사용자가 요청할 때만 한다(아래 "지적사항 반영" 참고).

이 워크플로우의 전제

구현 모델이 저장소를 뒤지느라 토큰을 쓰면 가성비가 사라진다. 탐색은 기획 단계에서 끝내고, 구현 단계는 코드만 쓰게 만드는 것이 설계 의도다. PLAN.md가 부실하면 이 워크플로우는 그냥 비싼 우회로가 된다. 계획서 품질에 가장 신경 써라.

언제 쓰지 말아야 하나

먼저 판단하고, 아니면 그냥 직접 해라. 사용자가 명시적으로 이 워크플로우를 지정했으면 그대로 따른다.

  • 한두 줄 수정, 오타, 설정값 변경 → 3단계 오케스트레이션 비용이 작업보다 크다.
  • 요구사항 자체가 아직 흐릿함 → 먼저 사용자와 정리해라. 계획서는 대화의 대체물이 아니다.
  • 탐색적 디버깅("왜 이게 안 되지") → 계획을 세울 수 있는 상태가 아니다. 원인부터 찾아라.

절차

0. 준비

codex CLI와 로그인 상태를 확인한다. 없으면 바로 알리고 멈춰라.

codex --version

작업 슬러그를 정하고 실행 디렉터리를 만든다. NNN.pnb/runs/ 아래 기존 항목 다음 번호(3자리, 001부터).

.pnb/runs/<NNN>-<영문-슬러그>/

.pnb/.gitignore에 없고 저장소가 git이면, 산출물을 커밋할지 사용자에게 한 번 묻고 원치 않으면 .gitignore.pnb/를 추가해라. 묻지 않고 임의로 .gitignore를 고치지 마라.

1. 기획 — Fable

fable-planner 서브에이전트를 띄운다. 플러그인으로 설치했으면 claude-codex:fable-planner, 저장소에 직접 복사했으면 fable-planner다. Agent 호출 시 model: "fable"을 명시해라(에이전트 정의에도 있지만 이중으로 보장한다).

프롬프트에 반드시 넣을 것:

  • 사용자의 원래 요청 원문 그대로. 요약해서 넘기지 마라 — 요약 과정에서 요구가 소실된다.
  • 이번 대화에서 사용자가 추가로 말한 제약(쓰지 말 라이브러리, 기한, 스타일 등).
  • PLAN.md를 쓸 절대 경로: <RUN_DIR>/PLAN.md
  • 저장소 루트 절대 경로.

에이전트가 돌아오면 반환값에서 OPEN_QUESTIONS를 본다.

  • 0 → 2단계로 진행. 진행 전에 사용자에게 계획 요약(제목, 건드릴 파일 수, 2~3문장 요약)과 PLAN.md 경로를 한 번 보여준다.
  • 1 이상 → 멈춘다. PLAN.md## 8. 확인 필요를 읽고 AskUserQuestion으로 물어라. 답을 받으면 그 답을 프롬프트에 얹어 fable-planner를 다시 호출해 계획을 갱신시킨다(같은 경로에 덮어쓰기). 구현자에게 미해결 질문이 담긴 계획을 넘기지 마라.

2. 구현 — codex-luna-max

먼저 <RUN_DIR>/BUILD_PROMPT.md를 만든다. 템플릿은 references/build-prompt.md에 있다. <PLAN_PATH> 자리에 PLAN.md의 절대 경로를 넣고, 그 외에는 문구를 바꾸지 마라.

그다음 스크립트를 백그라운드로 실행한다. 구현은 10분을 넘길 수 있어서 포그라운드 타임아웃에 걸린다. 백그라운드로 돌리면 종료 시 알림이 온다.

Read the full file on GitHub · 166 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 166 lines · 114 tokens per session scan A bfe0979f9266

Subscribe to this mod's changes

plan-and-build is a skill published in the GitHub repository newrise0410/claude-codex-workflow (4 stars, last pushed 1mo ago), licensed MIT. It adds 114 tokens to every session and 2,976 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.