Borrowing it
Nothing to install: this file belongs to LUKAWI/super-plumber. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/LUKAWI/super-plumber/master/.pi/skills/plumber-design/SKILL.mdgit clone --depth 1 https://github.com/LUKAWI/super-plumberWrote 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.
[](https://agentmods.dev/skills/lukawi/super-plumber/plumber-design)<a href="https://agentmods.dev/skills/lukawi/super-plumber/plumber-design"><img src="https://agentmods.dev/badge/skills/lukawi/super-plumber/plumber-design.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00152 | $0.03043 |
| Opus 5 | $0.00076 | $0.01522 |
| Sonnet 5 | $0.00030 | $0.00609 |
| Haiku 4.5 | $0.00015 | $0.00304 |
Grade A, and why
plumber-design 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plumber Design — 主线程编排剧本:需求 → 拓扑图 → 预览审核
Overview
- 产出物 = 用户审核通过的拓扑图(v0.5 起含领域结构:bounded context 与 ADR 是图中一等公民),不是执行结果。
- 本页是完整设计流程:无 sp-designer 也可独立完成需求理解、建图、验证与审核。Operations 手册提供操作语法;sp-designer 是可选的 subagent 角色适配,不是本技能的运行依赖。细则按需读 attachments/。
审核是硬性 gate:用户批准之前绝不进入执行阶段;审核通过后切换 plumber-execute。
代理层级边界:skill 文本和“有 subagent 工具”都不能可靠识别代理深度,故每个会话默认无派单权,不得从能力存在推断授权。只有直接承接用户需求的外层协调者可持一次性、不可转授的 orchestrator 授权直接派 sp-designer;授权在派发时耗尽,子任务只收到 leaf 约束,绝不可向下复制。缺授权一律走 solo。当前通用宿主未向 skill 暴露可验证的签名层级令牌,因此这是行为协议而非硬隔离:需要强保证的宿主还必须在工具层移除子 agent 的派发能力。
When to Use / 衔接关系(subagent 与 skill 已解耦)
- Use:接到新需求需拆解为 3–20 个有依赖关系的任务;设计或修改
.graph/拓扑(entry/exit、节点、边);验证拓扑质量;serve 预览并请求用户审核;审核后提修改意见 → 回本 skill(改图 → 重验 → 重预览 → 再审)。Do NOT:执行已审核通过的拓扑(用 plumber-execute);纯 todo 列表。 - 本技能保留完整设计方法;sp-designer 供支持 subagent 的应用独立派单使用,两者允许必要的方法重叠。
- 外层协调者持未使用的
orchestrator授权且需要专职设计 → 按「派单模板」直接派一个叶子sp-designer;未持授权或检测不到可用 subagent → 走「solo 分支」主线程扮演。被派会话一律不得再派。 - 用户审核通过 → 进入
plumber-execute(建议用户明确说"开始执行");节点状态裁决 / checkpoint 聚合 / 重试管理 →super-marioagent。
流程(Phase 0 定档 + 5 步,严格按序,绝不跳步)
Phase 0 — 路由决策表(两问定档,先于 Step 1;DEC-2,adr_0018 修订)
按顺序定档:① 关键未知是否阻止形成可信交付计划? 是 → program(切 chart 模式,研究后渐进完善计划);否 → ② 一个会话内能完成并验收吗? 是 → quick(单节点图,entry/exit 各一句话,跳过 serve 人审与 doctor、只跑 graph validate,claim→report→passed,init 后 graph approve --by <agent 名> --status self 自签);否 → standard(Step 1–5 一项不减)。跨会话、跨图、跨仓库和并行人数不单独触发 program。可信计划的定义、判例、存疑处理、转档条件与凭据纪律(adr_0016)→ attachments/workflow-classes.md;定档后出现使主要计划失效的关键未知 → 回本节重判。
Step 1 — 理解需求
向用户确认:目标是什么?交付物是什么?边界(不做什么)?有歧义就问,不要猜——需求不清,图必错。
Step 2 — 设计建图(可选委派)(先 entry/exit 后 L1-Ln 是铁律)
- 两条铁律写进派单:entry 和 exit 永远第一个定义(先图级字段后 L1-Ln 分层动脉);需求跨多个关注点 → 领域建模与 ADR 甄别是 designer 必做项(bounded context 划界/术语表/跨 context 契约边、够三判据的 ADR),点名不可裁剪。节点类型×默认纪律与 plan 纪律指针写法(WF07)→
attachments/disciplines-map.md。 - 按下列完整建图序列执行并通过体检,操作语法查手册 §2;不要求读取或安装 designer 角色文件。置层准则(IL-001):level 表达少数有意义的分层带(与分期/领域结构对齐),不把依赖深度编码进 level,层数建议 ≤5、避免单节点层;designer 出图前先声明置层方案(分几带、每带含义),声明与图不符按设计缺陷返工——细则 → 手册 §2.9。
- 完整建图序列:先定义图级 entry(输入/约束)与 exit(交付物/验收),再划 L1 主干(通常 3–7 项,quick 单节点例外)并细化可执行节点。每个 task 写明 plan、可核验 DoD、level 与 checkpoints(通常 2–4 个,quick 按档位裁剪)。依赖边 source 为前置、target 为后继;按真实门槛选 depends_on、fan_out/fan_in 等边型,不把所有关系都写成阻塞依赖(九边语义查手册 §3)。跨关注点时建立 context 的 boundary、glossary、contracts,明确成员归属与跨域契约。决策同时满足“难逆转、脱离上下文令人费解、存在真实权衡”才建 proposed ADR 并以 decides 挂到管辖对象,否则写进 plan;accept/supersede 留给裁决方。设计报告列 L1、置层方案、context/ADR 数量(为零说明理由)和验证结果。
- 查历史拒绝理由(WF05,出图/改图前必做):看软删归档(
ls .graph/*/nodes/*.deleted*.yaml,读deleted_reason等顶层键)→graph events -k node_deleted --json补全旧档理由 → 待建/待改节点命中已删方案时,设计报告逐一回应删除理由(换道,或给出理由已失效的依据),无回应 = 重蹈已否决方案,按设计缺陷返工。有 subagent 时主线程先筛、命中结果随派单交付 designer;solo 分支动笔前自跑。 - 有未使用的
orchestrator授权且需要专职设计:按「派单模板」直接派出一个叶子 sp-designer,等待其交付.graph/与设计报告;无授权、无 subagent 或当前已经被派 → 走「solo 分支」。
What ships with it
5 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.
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.
- yesterday Changed · +1 lines 3a62dcf7d03a
- 3d ago Changed · -42 lines e9219cf0067a
- 4d ago Changed · +22 lines 2be7d0080400
- 8d ago First seen · 92 lines · 152 tokens per session scan A b446342b08e0
plumber-design is a skill published in the GitHub repository LUKAWI/super-plumber (3 stars, last pushed yesterday), licensed MIT. It adds 152 tokens to every session and 3,043 once invoked, about $0.0008 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…