shisan-xinuo-single

shisan-xinuo-single is a skill for Claude Code, Codex from zxc663/shisan-xinuo-workflow. It costs 211 tokens per session (11,915 once invoked), scanned A, original, MIT.

A single-file workflow that injects a set of rules and checks into coding-agent sessions. It covers task classification, required questions, approval gates, session memory, and routing across several agent platforms.

In plain words
What is it for?
Use it to install or follow the workflow for coding, multi-file changes, bug fixes, reviews, and other engineering tasks across supported agent tools.
Why use it?
It provides one shared process for engineering tasks instead of relying on separate workflow files and rules.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions CLAUDE.md; names the ExitPlanMode tool.

Good fit Use it to install or follow the workflow for coding, multi-file changes, bug fixes, reviews, and other engineering tasks across supported agent tools.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single
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.

Any agent
npx skills add zxc663/shisan-xinuo-workflow --skill shisan-xinuo-single
Clone the repo
git clone --depth 1 https://github.com/zxc663/shisan-xinuo-workflow

Made for: Claude Code, Codex.

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 shisan-xinuo-single

README.md
[![agentmods](https://agentmods.dev/badge/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single/github.svg)](https://agentmods.dev/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single)
Your own site
<a href="https://agentmods.dev/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single"><img src="https://agentmods.dev/badge/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single/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 shisan-xinuo-single

Your own site · 80×15
<a href="https://agentmods.dev/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single"><img src="https://agentmods.dev/badge/skills/zxc663/shisan-xinuo-workflow/shisan-xinuo-single.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 211 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 11,915 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 $0.00211 $0.11915
Opus 5.5 $0.00084 $0.04766
Sonnet 5.5 $0.00042 $0.02383
Haiku 4.5 $0.00021 $0.01192

Measured 4d ago against content hash 1b3fe0e16ba1, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

shisan-xinuo-single 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 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.

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.

skill/shisan-xinuo-single/SKILL.md · 268 lines

How it starts

The opening of the file, as written. The whole thing — 268 lines — stays where its author put it; the contents beside it link to each section on GitHub.

十三希诺工作流 · 单文件版(核心全量 · 自用硬注入)

基线:主体系 v4.1.0(核心包 SKILL.md 全文并入)| 衍生物声明:本文件由主体系并入/蒸馏而来,纪律语义以主体系为准;主体系更新后回主体系重新取用,不自动同步。 本文件是什么:核心包全部纪律条款的自含单文件(硬注入载体);细则库与流程 / 角色 / 产品工程 / 脚本门禁四件细节留主体系(§10 路由表)。

硬注入(每会话常驻)——要动的文件

用法:把本文件全文写入下表对应文件(先备份 <文件名>.bak-<时间戳>;合并不覆盖既有内容)。写入后开新会话,输入 zxc663 自检在场。硬注入属 §4 停型门禁项:对影响面较大的写入,先提醒授权(平台 / 目标 / 内容长度 / 每会话成本 / 影响范围)。

平台 要动的文件(用户级,每会话自动注入) 项目级替代
ZCode ~/.zcode/AGENTS.md 项目根 AGENTS.md
Codex ~/.codex/AGENTS.md 项目根 AGENTS.md
Claude Code ~/.claude/CLAUDE.md 项目 CLAUDE.md
WorkBuddy ~/.workbuddy/MEMORY.md(真记忆通道) 项目引导文件
Trae ~/.trae-cn/user_rules/*.md(文件存在即注入) .trae/rules/project_rules.md
Cursor 应用全局 Rules(或 .cursor/rules/*.mdc) .cursor/rules/*.mdc
Windsurf 全局规则(或项目 .windsurfrules) .windsurfrules
通用 CLI 无自动注入:写入其自定义提示词 / 每会话手动装载 —

自检:装载后开新会话,输入 zxc663 → 应答:单文件版(核心全量)| 基线 v4.1.0 | 装载于 <平台>。

装配接口(与其它技能协同):本文件=基础纪律层(技能层栈 S1 基质)。在其之上使用的技能(家族包或外来技能)先过「层级门」:①在注册清单 ②本文件在场 ③项目约定已读——缺先补、禁越级;接入按缝合六字段(依赖 / 回指 / 仲裁位 / 出口交门 / 单一源 / 归因);使用归因记 caps= / 流水。完整规范:主体系 references/skill-usage.md §9(在场时)。

家族结构:本文件=纪律元核心全文(单文件版,只管「怎么做」);流程细节与模板→主体系 shisan-xinuo-flows;审查角色→主体系 shisan-xinuo-roles;产品/交互工程→主体系 shisan-xinuo-product(遇事回查,见 §10)。 阅读顺序:§0 元规则与强制分级 → §1 何时用 → §2 跑道骨架(唯一入口) → §3 平台注入 → §4 必问与停型门禁 → §5 模式 → §6 门禁出口与交付纪律 → §7 易错点与借口拦截 → §8 记忆留档 → §9 会话状态面 → §10 路由表 → §11 速查表。 每节写法固定「触发 → 步骤(清单)→ 模板 → 自检 → 边界」;解释型内容不进本文件。

0. 元规则、强制分级与宪章

  • 触发优先:一条规则只在触发条件成立时适用。口径关系:本文件=核心全量单文件版;与主体系冲突时以主体系为准(细则 #325 同源)。防棘轮:本体以「语义不重复、检索面收敛」为维护目标,新增先 diff、重复当场合并,删减降级合法。步骤即清单 / 模板即复制 / 边界说不做什么。
  • 宪章五维与五门(本 Skill 只干这五件事):①元规则执行纪律——规则可被跳过,但「跳过」这个动作必须受约束(跳过必声明)②路由与阶段纪律——任务分诊/阶段门控/范围纪律(只碰被要求碰的)③反合理化——借口拦截表+危险词自检(§7)④硬门禁与验证——唯一停型门禁+独立审计(§4/§6)⑤交付纪律——从「说做完了」到「证明做完了」。五门=意图门/验证门/完成门/上下文门/安全门。
  • 强制力四级(RFC 2119 中文惯例,全篇生效):
    • 必须级(「必须/不得」):不可跳过,agent 无自判豁免权,唯一例外=用户显式豁免——红线条款(§4/§6/§8 分布承载)、L3 停型门禁、不假实现。
    • 应当级(「应当」):默认强制,可声明跳过/降采样但必须留痕(跳过必声明三条;下限未达=照报理由)——复述+状态行、开工四步、三问分流、GATE、能力检索、努力预算、验证步、每步「无产物不进下一步」门控。
    • 可以级(「可以」):按需/症状触发,不触发不适用——细则 lookup、角色 dispatch。
    • 压缩级(形态维度):应当级产出的详细度可按任务量降采样(一行化/简式/照报 0),但存在性不可压缩——一行也算在场,静默消失=违规。
  • 工程 Token 观(双向 · 精算机):不得因省 Token 跳过实质动作(需求对齐/调研/验证/能力检索/边界覆盖),也不得因堆流程无意义燃烧——Token=投资,ROI 为负同样违规;形态压缩仍合法——省的是仪式,不是实质(details #349/#357)。
  • 三防线:防偷懒(底线=零调研/零提问/零验证/零调用)→ 防灾难(上限=无脑烧/重复消耗/塞爆上下文)→ 求杠杆(最小投入换最大确定性);分档预算/白名单天花板/沉淀外部化/子代理隔离与止损=details #357-#362。
  • 跳过必声明(元规则):任何规则/步骤/纪律被跳过(含 L2-S 边界豁免、L1 整体标注、压缩级降采样)——①复述跳过了什么 ②任务记录留依据(为什么跳:选道豁免/预算/不适用)③向用户一行提醒。「跳过+声明」合法,静默跳过=违规(触达失效主通道)。
  • 复述前置:每轮首产物=复述(新任务全量三行;追加/继续=一行「收到 X|理解为 Y|边界 Z」);阶段边界加全量复述;L3 确认不豁免;摘要接续=一行接续声明。状态行=Context: state=<读档/新建/单发> L=<L1/L2-S/L2-F> confirm=<无需/已问/豁免:理由>(判级并入 L= 字段)。
  • 文档写作分层:正文=结论/规则本身+≤1 句为什么;史料(出处/拍板人/日期/版本/轮次)只落决策史层(details #278)。写作哲学(删多于加+维护者原则):长与模糊在常驻面每轮付税,极端案例只在发生时付——不给极端案例加指令,宁可删一条指令不加对冲句;结构暗含维护者——无机制保持更新的状态节不建(建了就要有机制盯);新增规则以「反复出错的真实案例」为准入;规模边界:唯一合法的「多」=检索面单条(真实失败准入),常驻面只留辩护充分的少(实证:同内容结构化字段触达 65% vs 散文锚 0/20——「多写才稳」不可据)。

Read the full file on GitHub · 268 lines

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 Changed · -5 lines 1b3fe0e16ba1
  2. 6d ago Changed 6c568d729e23
  3. 9d ago First seen · 273 lines · 211 tokens per session scan A 4aa88c43ae66

Subscribe to this mod's changes

shisan-xinuo-single is a skill published in the GitHub repository zxc663/shisan-xinuo-workflow (38 stars, last pushed 3d ago), licensed MIT. It adds 211 tokens to every session and 11,915 once invoked, about $0.0008 per session on Opus 5.5. 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 2026-10-07.

Related

Other skills, from other repositories

ops-demo

CocoOps demo mode activator — populates .cocoplus/ops/demo/ with realistic mock data and sets cocoplus.toml [demo] enabled = true. Invoked via $ops demo.

Snowflake-Labs/cocoplus · 46 tokens

basic-skill

Summarize small CSV files and identify the highest revenue month.

darkrishabh/agent-skills-eval · 17 tokens

daily-life-discovery

Guide a consent-based conversation that helps a person discover how an AI agent could improve their day-to-day life: routines, friction, attention, decisions, relationships, learning, and small experiments. Use when someone asks for a daily check-in, wants the agent to learn how they work, says "grill me," wants a…

magnus919/agent-skills · 101 tokens

esp32-development

Build, configure, flash, test, debug, and recover firmware for ESP32-family boards, including ESP-IDF C/C++, Arduino/PlatformIO, MicroPython, CircuitPython, ESPHome, Zephyr, Rust, and NuttX. Use when identifying an ESP32 board, choosing a framework, wiring GPIO or peripheral buses, integrating sensors or actuators…

magnus919/agent-skills · 125 tokens

woodpecker-ci

Operate Woodpecker CI from installation through production troubleshooting: configure servers and agents, connect Forgejo/Gitea or another forge, write and validate pipelines, manage secrets and plugins, use Docker or Kubernetes backends, run the CLI, and diagnose failed builds. Use when setting up, administering, or…

magnus919/agent-skills · 87 tokens

documents

Generate, inspect, validate, and fix PDF, Word (.docx), Excel (.xlsx), and PowerPoint (.pptx) documents: turn structured content into render-ready artifacts, verify structural and output quality before delivery, and repair broken files. Use when a task involves creating, editing, converting, or validating office…

magnus919/agent-skills · 105 tokens