subagent-driven-development

subagent-driven-development is a skill for Claude Code, Codex from jnMetaCode/superpowers-zh. It costs 21 tokens per session (7,929 once invoked), scanned C, original, MIT.

A development workflow in which a fresh coding agent implements each independent task, another review checks that task, and a final review examines the complete set of changes.

In plain words
What is it for?
Use it when carrying out a plan with several mostly independent coding tasks in the same session, especially when each task can be implemented and reviewed separately.
Why use it?
It separates task context and adds checks for both requirement compliance and code quality. This helps catch missed details during a multi-step implementation plan.

Skill for Claude CodeCodex

Part of the superpowers-zh plugin — 20 skills, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/jnmetacode/superpowers-zh/subagent-driven-development
Any agent
npx skills add jnMetaCode/superpowers-zh --skill subagent-driven-development
Clone the repo
git clone --depth 1 https://github.com/jnMetaCode/superpowers-zh

Made for: Claude Code, Codex.

Or install superpowers-zh, the plugin that ships this one along with the rest of its 20 skills, 1 hook.

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 subagent-driven-development

README.md
[![agentmods](https://agentmods.dev/badge/skills/jnmetacode/superpowers-zh/subagent-driven-development.svg)](https://agentmods.dev/skills/jnmetacode/superpowers-zh/subagent-driven-development)
Your own site
<a href="https://agentmods.dev/skills/jnmetacode/superpowers-zh/subagent-driven-development"><img src="https://agentmods.dev/badge/skills/jnmetacode/superpowers-zh/subagent-driven-development.svg" alt="Measured on agentmods" height="20"></a>
Per session 21 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,929 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. Scan, not verified.
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 $0.00021 $0.07929
Opus 5 $0.00010 $0.03964
Sonnet 5 $0.00004 $0.01586
Haiku 4.5 $0.00002 $0.00793

Measured 4d ago against content hash 27081029a6b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

subagent-driven-development scanned grade C with 1 finding 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 4d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

当覆盖整个分支的最终审查干净、且它的修复已合并时,删除**本计划**的工作区(`rm -rf <工作区>`)——现在 git 历史就是记录了。同级目录属于别的计划,别去动它们。
skills/subagent-driven-development/SKILL.md · 342 lines

How it starts

The opening of the file, as written. The whole thing — 342 lines — stays where its author put it; the contents beside it link to each section on GitHub.

子智能体驱动开发

通过为每个任务分派一个全新的实现子智能体来执行计划:每个任务完成后做一次任务审查(规格合规性 + 代码质量),全部任务结束后再做一次覆盖整个分支的宽范围审查。

为什么用子智能体: 你把任务委派给具有隔离上下文的专用智能体。通过精心设计它们的指令和上下文,确保它们专注并成功完成任务。它们绝不应继承你会话的上下文或历史记录——你要精确构造它们所需的一切。这样也能为你自己保留用于协调工作的上下文。

核心原则: 每个任务一个全新子智能体 + 任务审查(规格 + 质量)+ 结尾宽范围审查 = 高质量、快速迭代

旁白: 工具调用之间最多说一句简短的旁白——进度账本和工具结果本身就是记录。

持续执行: 不要在任务之间停下来向你的人类伙伴确认。不间断地执行计划里的所有任务。唯一该停下的理由是:你无法解决的 BLOCKED 状态、确实妨碍推进的歧义,或所有任务已完成。"我该继续吗?"之类的询问和进度小结都在浪费他们的时间——他们让你执行计划,那就执行。

何时使用

digraph when_to_use {
    "有实现计划?" [shape=diamond];
    "任务基本独立?" [shape=diamond];
    "留在当前会话?" [shape=diamond];
    "subagent-driven-development" [shape=box];
    "executing-plans" [shape=box];
    "手动执行或先头脑风暴" [shape=box];

    "有实现计划?" -> "任务基本独立?" [label="是"];
    "有实现计划?" -> "手动执行或先头脑风暴" [label="否"];
    "任务基本独立?" -> "留在当前会话?" [label="是"];
    "任务基本独立?" -> "手动执行或先头脑风暴" [label="否 - 紧密耦合"];
    "留在当前会话?" -> "subagent-driven-development" [label="是"];
    "留在当前会话?" -> "executing-plans" [label="否 - 并行会话"];
}

与 Executing Plans(并行会话)的对比:

  • 同一会话(无上下文切换)
  • 每个任务全新子智能体(无上下文污染)
  • 每个任务后做审查(规格合规性 + 代码质量),结尾做宽范围审查
  • 更快的迭代(任务间无需人工介入)

流程

digraph process {
    rankdir=TB;

    subgraph cluster_per_task {
        label="每个任务";
        "分派实现子智能体 (./implementer-prompt.md)" [shape=box];
        "实现者有疑问?" [shape=diamond];
        "回答问题,提供上下文" [shape=box];
        "实现者实现、测试、提交、自审" [shape=box];
        "生成审查包,分派任务审查者 (./task-reviewer-prompt.md)" [shape=box];
        "规格 ✅ 且质量通过?" [shape=diamond];
        "发现与计划原文冲突?" [shape=diamond];
        "询问人类伙伴以哪个为准" [shape=box];
        "第 R/5 轮修复: R≤3 唤回原实现者; R≥4 换全新实现者 + 更强模型" [shape=box];
        "分派定向复审 (./re-review-prompt.md)" [shape=box];
        "所有发现都已解决?" [shape=diamond];
        "R = 5?" [shape=diamond];
        "逐条裁定未解决的发现" [shape=box];
        "存在承重的发现?" [shape=diamond];
        "停止: 向人类伙伴报告 BLOCKED" [shape=box];
        "把发现连同裁定搁置进账本" [shape=box];
        "往账本追加完成行,标记待办完成" [shape=box];
    }

    "准备: 工作树、查账本、读计划、起飞前审查" [shape=box];
    "还有任务?" [shape=diamond];
    "分派最终代码审查者 (../requesting-code-review/code-reviewer.md)" [shape=box];
    "最终审查有发现? 一次修复分派、一次定向复审、裁定残留项" [shape=box];
    "最终审查干净: 删除本计划的工作区" [shape=box];
    "使用 superpowers:finishing-a-development-branch" [shape=box style=filled fillcolor=lightgreen];

    "准备: 工作树、查账本、读计划、起飞前审查" -> "分派实现子智能体 (./implementer-prompt.md)";
    "分派实现子智能体 (./implementer-prompt.md)" -> "实现者有疑问?";
    "实现者有疑问?" -> "回答问题,提供上下文" [label="是"];
    "回答问题,提供上下文" -> "实现者实现、测试、提交、自审";
    "实现者有疑问?" -> "实现者实现、测试、提交、自审" [label="否"];
    "实现者实现、测试、提交、自审" -> "生成审查包,分派任务审查者 (./task-reviewer-prompt.md)";
    "生成审查包,分派任务审查者 (./task-reviewer-prompt.md)" -> "规格 ✅ 且质量通过?";
    "规格 ✅ 且质量通过?" -> "往账本追加完成行,标记待办完成" [label="是"];
    "规格 ✅ 且质量通过?" -> "发现与计划原文冲突?" [label="否"];
    "发现与计划原文冲突?" -> "询问人类伙伴以哪个为准" [label="是"];
    "询问人类伙伴以哪个为准" -> "第 R/5 轮修复: R≤3 唤回原实现者; R≥4 换全新实现者 + 更强模型";
    "发现与计划原文冲突?" -> "第 R/5 轮修复: R≤3 唤回原实现者; R≥4 换全新实现者 + 更强模型" [label="否"];
    "第 R/5 轮修复: R≤3 唤回原实现者; R≥4 换全新实现者 + 更强模型" -> "分派定向复审 (./re-review-prompt.md)";
    "分派定向复审 (./re-review-prompt.md)" -> "所有发现都已解决?";
    "所有发现都已解决?" -> "往账本追加完成行,标记待办完成" [label="是"];
    "所有发现都已解决?" -> "R = 5?" [label="否"];
    "R = 5?" -> "第 R/5 轮修复: R≤3 唤回原实现者; R≥4 换全新实现者 + 更强模型" [label="否 - 进入下一轮"];
    "R = 5?" -> "逐条裁定未解决的发现" [label="是 - 熔断触发"];
    "逐条裁定未解决的发现" -> "存在承重的发现?";
    "存在承重的发现?" -> "停止: 向人类伙伴报告 BLOCKED" [label="是"];
    "存在承重的发现?" -> "把发现连同裁定搁置进账本" [label="否"];
    "把发现连同裁定搁置进账本" -> "往账本追加完成行,标记待办完成";
    "往账本追加完成行,标记待办完成" -> "还有任务?";
    "还有任务?" -> "分派实现子智能体 (./implementer-prompt.md)" [label="是"];
    "还有任务?" -> "分派最终代码审查者 (../requesting-code-review/code-reviewer.md)" [label="否"];
    "分派最终代码审查者 (../requesting-code-review/code-reviewer.md)" -> "最终审查有发现? 一次修复分派、一次定向复审、裁定残留项";
    "最终审查有发现? 一次修复分派、一次定向复审、裁定残留项" -> "最终审查干净: 删除本计划的工作区";
    "最终审查干净: 删除本计划的工作区" -> "使用 superpowers:finishing-a-development-branch";
}

Read the full file on GitHub · 342 lines

Files

What ships with it

6 files 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. 4d ago First seen · 342 lines · 21 tokens per session scan C 27081029a6b0

Subscribe to this mod's changes

subagent-driven-development is a skill published in the GitHub repository jnMetaCode/superpowers-zh (7,965 stars, last pushed yesterday), licensed MIT. It adds 21 tokens to every session and 7,929 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.