团队 AI Native 协作操作说明(团队成员版)

给产品、交付、开发、测试、项目经理、售前等团队成员看:这个项目怎么用、资料怎么找、任务怎么输入、skill / workflow / HTML review 页面怎么复用。管理者日常控制台见 AI_NATIVE_PROJECT_CONTROL_CONSOLE.html,领导版成果介绍见 AI_AGENT_PROJECT_INTRO_FOR_BEGINNERS.html

团队成员版 产品类 + 项目类 企业微信微盘索引 skill / workflow / HTML 2026-07-23
文档版本V1.1
更新日期2026-07-23
数据快照2026-07-23
本次更新复用导航、资产口径、代表性落地、版本履历

0.1 新增:《创造》产品创造方法组

员工版方法组 HTML

把《创造:用非传统方式做有价值的事》蒸馏成 16 个产品、创新、组织和经营 skill。

知识包入口

查看整书理解、应用手册、候选项、拒绝项、验证报告和来源边界。

OSS 可转发链接

发给同事时优先用这个外网 HTML 链接;项目内源码仍保存在 Git。

0.2 当前资产快照:团队可以直接复用什么

下面是 2026-07-23 的本地项目快照。数量说明资产已经形成,不代表每项都已被全员采用;效率、收入、毛利和客户验收仍要看对应任务证据。

212 个项目 skill · 148 个流程资产

遇到重复、易错、需要停止规则的工作,先让 Agent 从 项目 skill 目录 自动路由,不要靠新人背名称。

326 条任务证据 · 60 份团队线程记录

恢复任务先读 QUEUEHEARTBEAT、任务索引和线程记录,不把聊天结论当完成证据。

59,473 条文件索引 · 7,209 条文档索引

找公司/项目资料先从 公司知识库入口 定位,原始大文件仍留在微盘或受控源目录。

570 个 HTML · 309 份 summary · 199 份 Wiki

Markdown 是 Agent 维护真源;HTML 给人审、展示和培训;summary 和 Wiki 负责恢复上下文与复用判断。

八类高价值沉淀怎么用

沉淀你什么时候用真实入口边界
控制与治理新任务归类、跨仓库协作、状态恢复README · IPO 指南 · control控制项目不承载业务代码
产品与战略产品规划、场景、软件、硬件、AI、交付资产modules/product · 智慧食堂战略策略文档不是已验证经营结果
知识与索引检索微盘、项目资料、方法论和历史决策company knowledge · LLM Wiki索引不替代原文复核
Skill 与 workflow同类任务重复出现、容易漏步骤或越权skill registry一次性说明不要伪装成 skill
开发与测试需求落代码、接口、测试、提交和回滚开发 Agent 控制台 · harness测试通过不等于生产发布成功
交付与手册需要人审、培训、用户操作或客户交付操作手册入口 · 交付资产化对外材料逐件脱敏和确认
团队学习记忆会议、问答、反馈和复盘需要团队继续使用线程记录规则被否定做法只记反模式,不沉淀成正向经验
专业 Agent代码评审、开发、项目调度、售前、知识库、阿里云运维等专业协作Agent Control Room · 阿里云运维 Agent登记不等于全部已独立部署;高风险动作仍人工审批

近期代表性落地:先看证据,再决定怎么复用

阿里云运维 Agent V0

已验证独立运行、真实资产盘点、计划审批和生产执行阻断;适合学习高风险 Agent 的最小权限与人工门禁。

产品交付双泳道

把 REQ 需求与开发交付、DEP 使用与部署交付分开,避免“代码完成”被误报为“客户环境已交付”。优先级仍待人工确认。

企业/学校营养产品战略 V0.1

展示如何把产品事实、市场判断、证据等级和待验证指标放在同一页;战略建议不是经营结果。

应急安全整改证据包

展示如何把漏洞事实、影响、整改计划、复测与责任边界分开;材料形成不等于线上整改已全部闭环。

0. 先找之前那份 HTML 在哪里

原始审阅 HTML:

work/2026-05-21-ai-agent-project-introduction/review.html

已复制到顶层的人读入口:

AI_AGENT_PROJECT_INTRO_FOR_BEGINNERS.html

稳定 Markdown 真源:

standards-stack/ai-engineering/AI_AGENT_PROJECT_INTRO_FOR_BEGINNERS.md

原来的 HTML 是一次工作证据包里的 review 页面,已经复制到顶层方便给领导看成果。本文件 TEAM_AI_NATIVE_COLLABORATION_GUIDE.html 是团队成员操作说明;Markdown 真源留在 work/2026-05-25-team-ai-native-collaboration-guide/

1. 这个项目到底是什么

zhctprompt 不是业务代码仓库,而是产品开发部的 AI 协作控制项目。

它把产品资料、项目资料、企业微信微盘索引、云效任务、代码仓库、手册视频、复盘证据和 Agent 规则放到同一套可恢复的工作系统里。

以前:人问一句,AI 答一句,答完就散了。
现在:人给任务,Agent 先找资料、判断归类、执行、验证、留证据、沉淀成下次能复用的流程。

2. 新人看完要达到什么效果

过去现在
群里说一句,资料靠人转发先把目标、来源、范围、验收标准写清楚
微盘、云效、代码、文档各找各的modules/work_company_knowledge/tasks/ 统一进入
AI 每次都像第一次接触项目Agent 先读规则、索引、历史证据,再执行
做完只剩口头结论留下 Markdown、HTML、CSV、代码提交、测试、手册、视频或任务索引

3. 第一天下手顺序

顺序文件你要看懂什么
1README.mdzhctprompt 是控制项目,不是业务代码仓库
2TEAM_AI_NATIVE_COLLABORATION_GUIDE.html团队 AI native 协作怎么用
3PROJECT_IPO_USAGE_GUIDE.mdInput -> Process -> Output 和“一总五分”
4modules/README.md产品类和项目类资料怎么分
5work_company_knowledge/README.md企业微信微盘资料怎么检索

4. 总体结构:项目类 + 产品类

产品类:modules/product/

产品需求、产品战略、UI/原型、软件入口、开发项目资料入口。主要给产品、UI、方案、研发负责人使用。

项目类:modules/project/

企业微信微盘里的客户项目、历史项目、主题资料索引。主要给项目经理、交付、售前、产品、管理者使用。

公司微盘知识库:work_company_knowledge/

只存索引、分类、导航和快照,不批量复制原始文件。

项目证据:work_<project>/

单个项目的控制资料、方案、排查、证据和复盘,比如 work_store/work_qc/

5. 产品类目录怎么用

要做的事进入哪里输出什么
产品需求、流程、验收标准modules/product/requirements/需求说明、业务流程、能力地图、验收口径
UI、蓝湖、设计规范、原型modules/product/ui/UI 索引、设计规范、原型入口、截图证据
PC/移动端/小程序/API 入口梳理modules/product/software-platformization/软件入口地图、页面/接口能力说明
产品战略、产品地图、路线图modules/product/strategy/产品战略、SPAN/PDC/7-2-1、路线图、证据缺口
开发项目资料入口modules/product/development/storeai_apiai_storeai_app 等项目入口
产品评审modules/product/product-review-workbench/多人打分、评审结论、分歧风险、JSON 导入导出
产品人的规则:

客户案例、上线状态、收入、毛利、设备数量、验收结果必须有证据。没有证据就写 待确认,不要补编。

6. 项目类目录怎么用

要查什么入口
某个客户项目或主题是否有资料modules/project/wedrive-projects/project-index.tsv
微盘知识总览work_company_knowledge/knowledge-map.md
全量源文件路径work_company_knowledge/indexes/file-inventory.tsv
Word/PDF/PPT/Excel/Markdown 等文档work_company_knowledge/indexes/document-index.tsv
企业微信在线文档/在线表格快照work_company_knowledge/online-documents/online-document-index.csv
客户/项目:
当前阶段:
资料来源:
业务场景:
涉及角色:
软件入口:
硬件触点:
验收指标:
风险和缺口:

7. 企业微信微盘怎么配合

企业微信微盘仍然是原始资料真源。本项目只保存派生索引和少量可读快照。

微盘来源本项目入口
项目交付管理共享空间work_company_knowledge/indexes/project-index.tsvdocument-index.tsv
数字技术中心知识库work_company_knowledge/knowledge-map.mdrole-entrypoints.tsv
企业微信在线文档/表格work_company_knowledge/online-documents/online-document-index.csv

8. 各个顶层目录怎么用

目录/文件谁用用法
README.md所有人新手总入口,先理解项目定位和本地启动方式
AGENTS.mdAgent / 高阶使用者执行规则、停止规则、项目边界
PROJECT_IPO_USAGE_GUIDE.md所有人用 Input -> Process -> Output 组织任务
HEARTBEAT.md接手人 / Agent最近做到哪、下一步是什么
tasks/QUEUE.md项目经理 / Agent当前 Ready、Blocked、Done 队列
control/项目经理 / Agent路径、状态、任务索引、验证脚本
modules/product/产品 / UI / 方案产品需求、UI、原型、战略、软件入口
modules/project/项目 / 交付 / 售前微盘项目和主题索引入口
work_company_knowledge/全员企业微信微盘派生知识库
work_<project>/项目负责人 / Agent项目专属控制资料和证据
standards-stack/管理者 / Agent治理规则、skills、AI 工程化、产品开发部标准
docker/开发 / 测试本地智慧餐厅 Docker 启动和验证

9. AI 协作标准流程

DEFINE
PLAN
BUILD
VERIFY
REVIEW
SHIP
阶段要做什么留下什么
DEFINE把自然语言需求变成目标、范围、非目标、验收标准任务契约或说明
PLAN拆最小切片,确定目录、仓库、分支、验证命令计划和证据路径
BUILD做文档、检索、代码、方案、手册或索引文件改动或产物
VERIFY跑测试、截图、解析、链接检查、格式检查验证结果
REVIEW检查 diff、风险、证据、人工 review 页面Markdown 真源和 HTML review
SHIP更新队列、心跳、任务索引、复用规则可恢复入口

10. 团队成员要会用的三类资产

截至 2026-07-23,standards-stack/agent-skills/skills/ 实际共有 212 个 SKILL.md,其中包含项目自有 skills、PM Skills 和团队路由 skill。团队执行时始终以实际目录和路由规则为准。

212 个团队 skill

原有项目规则、经验、停止条件和执行路径继续保留;PM Skills 额外提供 PRD、战略、研究、增长和上线检查流程。

流程类资产

团队的工作流骨架,把需求进入、计划、执行、验证、评审、交付、复盘固定下来。

HTML 页面

给人看的 review / 展示 / 培训界面;Markdown 仍然是 Agent 维护真源。

分工记忆:

Markdown 给 Agent 和后续维护人读;HTML 给人 review、展示、培训和交接;skill / workflow 给下次同类任务复用。

10.1 PM Skills 产品经理工作流

已经把开源 pm-skills 整套产品经理工作流装进本项目:9 个插件、68 个可安装 skill、42 个上游 command 源文件,并新增团队路由 skill pm-skills-team-workflow-router

快速出 PRD

用 PRD、用户故事、验收标准、风险和上线检查流程,把零散需求变成可评审文档。

战略 / 研究 / 增长

覆盖 strategy canvas、SWOT、PESTLE、Porter、Ansoff、用户研究、竞品、GTM、ICP 和 North Star。

AI 上线检查

用 AI Shipping Kit 做意图文档、测试覆盖、安全、性能和 ship packet 检查。

团队详细说明:

PM Skills 团队工作流使用指南 · 路由 skill · 导入摘要

请用 PM Skills 的 PRD 流程,针对下面需求快速出一版 PRD。
请用 PM Skills 做一次产品战略跨步,输出 strategy canvas、SWOT、PESTLE、Porter、Ansoff 和 30 天验证动作。
请用 PM Skills 的 AI Shipping Kit 检查这个应用,输出测试覆盖、安全、性能和上线 blocker。

10.2 阿里云运维 Agent 怎么协作

阿里云运维 Agent 是 zhctprompt 控制的首个独立专业 Agent。当前 V0 已部署并在线,但只开放资产盘点、只读诊断、变更计划和审计;生产重启、快照、任意 Shell / PowerShell 和自动修复仍然关闭。

告警进入
资产定位
只读诊断
精确审批
受控执行
业务回读
当前可用:只读控制面

真实盘点华北 2 的 21 台 ECS,其中 17 台云助手在线;服务由 systemd 托管,只监听 127.0.0.1:18765

当前验证:安全门禁有效

本地自动化测试 14 passed;计划创建 200、审批 200、生产执行 403,说明审批不能绕过环境和 RAM 权限。

当前边界:不自动重启

重启、快照、自动修复、任意命令和公网 API 尚未开放;必须先完成非关键实例演练和人工授权。

谁负责什么

角色必须完成不能越过的边界
告警发起人 / 项目负责人提供告警时间、实例 ID、业务影响、维护窗口和验证联系人。不能用“继续”“确认”代替具体变更授权。
Agent盘点资产、收集只读证据、生成带哈希、有效期、回退和验收的计划。不能自行决定重启生产服务器。
生产审批人核对实例、动作、窗口、影响、回退和验收,精确确认计划。不能把审批等同于已经具备云端执行权限。
技术执行人只在临时最小权限和实例白名单生效后执行计划内动作。不能运行计划外的任意 Shell / PowerShell。
业务验证人检查服务、页面、接口、任务和告警是否恢复,并回填证据。不能只看服务器在线或 HTTP 200 就宣布业务恢复。

员工发起时直接复制

请用阿里云运维 Agent 处理下面告警,先做只读诊断,不执行重启:

告警时间:
地域:
实例 ID / 实例名称:
内网 / 公网 IP:
业务影响:
允许维护窗口:
希望验证的服务或页面:
不能做的事:不重启、不删除文件、不修改安全组、不运行计划外命令。

请输出:资产定位、诊断证据、根因假设及证据等级、
变更计划草案、风险、回退和验收清单。
生产变更必须另起一步:

诊断完成后只能先生成变更计划。计划必须写清精确实例、动作、影响、快照/备份、执行窗口、业务验证、回退、计划哈希、有效期、审批人和执行人;普通聊天中的“确认”不构成执行授权。

工作入口: 阶段摘要 · 验收结果 · 人审页面 · 运行手册 · 本机控制脚本

11. 知识库、项目库和团队入口总导航

这块把团队最常用的 HTML、Markdown 知识库、项目库、微盘派生索引和 skill 安装入口集中挂出来。原则是:能点到真实文件;原始大文件仍留在微盘或本机源目录,项目里只放索引和派生成果。

团队 HTML 入口

团队成员总入口
新人培训、团队协作入口、skill / 知识库导航。

管理者控制台
管理者查看项目入口、资料、队列和绩效证据。

领导版成果介绍
给领导或非技术同事解释 AI Agent 项目价值。

开发 Agent 控制台
研发、测试、代码任务入口。

PM Skills 使用指南
68 个 PM skills 和上游 command 工作流说明。

阿里云运维 Agent
查看部署状态、验收证据、权限边界和后续开放条件。

Store 提交前 QA 闸门
报表聚合、消费策略、订单快照、退款状态机、权限和 Docker 环境提交前检查。

操作手册入口
用户手册、操作视频和培训材料入口。

保康技术材料
客户项目技术资料预览。

核心知识库

LLM Wiki 总入口
所有可复用方法论和研究知识入口。

产品方法库
产品经理方法、需求四层、王慧文产品课、卖点买点等。

需求四层知识页
需求证据深浅判断。

王慧文产品课
需求、市场、PMF、产品经理基本功。

卖点到买点
售前话术和客户价值翻译。

AI 组织方法论
AI 原生组织、RACI、对象模型和交付闭环。

AI Native 交付协同
项目交付、会议、验收和知识回流。

云效 Codex 协作
云效、Codeup、微盘、Codex 的协作闭环。

Store QA 近期缺陷复盘规则
近两周 store 缺陷学习、防复发约束和 Codex 响应触发规则。

团队 skill 与安装交接

Agent Skills README
项目 skills 的总说明和触发规则。

团队 skill 交接说明
复制给同事、安装到本机 Codex 的操作说明。

团队 skill Manifest CSV
团队 skill 清单,适合表格查看。

项目 skill 安装脚本
同事本机安装项目 skills 的脚本。

PM Skills 导入摘要
PM Skills 导入和使用边界。

旧版 skill 使用盘点
早期 101 个核心 skill 使用清单。

项目库 / 产品库

模块总入口
产品、项目、开发模块总入口。

产品模块入口
产品方案、UI、战略、开发项目资料入口。

项目模块入口
企业微盘项目索引入口。

微盘项目索引
企业微信微盘项目索引。

公司知识库
企业微信微盘派生知识库入口。

公司知识地图
按角色/主题找公司资料。

在线文档索引
在线文档、表格、接口导入记录。

源文件根目录
微盘和本机资料源定位。

Store QA 闸门 Markdown
Agent 提交前回填模板和专项验证矩阵。

资产组入口路径怎么用
团队 HTML 入口团队成员总入口TEAM_AI_NATIVE_COLLABORATION_GUIDE.html新人培训、团队协作入口、skill / 知识库导航。
团队 HTML 入口管理者控制台AI_NATIVE_PROJECT_CONTROL_CONSOLE.html管理者查看项目入口、资料、队列和绩效证据。
团队 HTML 入口领导版成果介绍AI_AGENT_PROJECT_INTRO_FOR_BEGINNERS.html给领导或非技术同事解释 AI Agent 项目价值。
团队 HTML 入口开发 Agent 控制台DEVELOPMENT_AGENT_CONSOLE.html研发、测试、代码任务入口。
团队 HTML 入口PM Skills 使用指南pm-skills-team-workflow-guide.html68 个 PM skills 和上游 command 工作流说明。
团队 HTML 入口阿里云运维 Agent 人审页work_aliyun_ops_agent/aliyun-ops-agent-v0-review.html查看部署状态、验收证据、权限边界和后续开放条件。
团队 HTML 入口Store 提交前 QA 闸门work_store/qa/store-pre-submit-qa-gate-checklist.htmlstore 提交前覆盖报表聚合、消费策略、订单快照、退款状态机、权限和 Docker 环境的可勾选检查清单。
团队 HTML 入口操作手册入口operation-manuals/index.html用户手册、操作视频和培训材料入口。
团队 HTML 入口保康技术材料product_kezuozhongqi-baokang-middle-school-smart-canteen/baokang-smart-canteen-technical-materials.html客户项目技术资料预览。
核心知识库LLM Wiki 总入口standards-stack/llm-wiki/README.md所有可复用方法论和研究知识入口。
核心知识库产品方法库standards-stack/llm-wiki/product-methods/README.md产品经理方法、需求四层、王慧文产品课、卖点买点等。
核心知识库需求四层知识页standards-stack/llm-wiki/product-methods/requirement-four-layer-method.md需求证据深浅判断。
核心知识库王慧文产品课standards-stack/llm-wiki/product-methods/wang-huiwen-qinghua-product-course.md需求、市场、PMF、产品经理基本功。
核心知识库卖点到买点standards-stack/llm-wiki/product-methods/selling-points-to-buying-points.md售前话术和客户价值翻译。
核心知识库AI 组织方法论standards-stack/llm-wiki/ai-organization-methodology/README.mdAI 原生组织、RACI、对象模型和交付闭环。
核心知识库AI Native 交付协同standards-stack/llm-wiki/ai-native-project-delivery-collaboration/README.md项目交付、会议、验收和知识回流。
核心知识库云效 Codex 协作standards-stack/llm-wiki/ai-yunxiao-codex-collaboration/README.md云效、Codeup、微盘、Codex 的协作闭环。
核心知识库思维模型库standards-stack/llm-wiki/thinking-models/README.md分析、规划、复盘、决策问题的模型路由。
核心知识库健康经济研究standards-stack/llm-wiki/health-economy-sports-research/README.md健康、体育、营养相关市场研究。
核心知识库B2B 销售预测 AIstandards-stack/llm-wiki/sales/b2b-sales-forecast-ai.md智慧食堂商机、漏斗和预测方法。
核心知识库Store QA 近期缺陷复盘规则work_store/qa/store-recent-bug-learning-guardrails-20260611.mdstore 近两周缺陷学习、防复发约束和 Codex 响应触发规则。
团队 skill 与安装交接Agent Skills READMEstandards-stack/agent-skills/README.md项目 skills 的总说明和触发规则。
团队 skill 与安装交接团队 skill 交接说明standards-stack/agent-skills/CODEX_TEAM_SKILL_HANDOFF.md复制给同事、安装到本机 Codex 的操作说明。
团队 skill 与安装交接团队 skill Manifest CSVstandards-stack/agent-skills/CODEX_TEAM_SKILL_MANIFEST.csv团队 skill 清单,适合表格查看。
团队 skill 与安装交接项目 skill 安装脚本standards-stack/agent-skills/scripts/install_project_skills_to_codex.sh同事本机安装项目 skills 的脚本。
团队 skill 与安装交接PM Skills 导入摘要work/2026-06-11-pm-skills-team-workflow/summary.mdPM Skills 导入和使用边界。
团队 skill 与安装交接旧版 skill 使用盘点work/2026-06-09-beginner-skill-usage-guide/project-skill-usage-inventory.csv早期 101 个核心 skill 使用清单。
项目库 / 产品库模块总入口modules/README.md产品、项目、开发模块总入口。
项目库 / 产品库产品模块入口modules/product/README.md产品方案、UI、战略、开发项目资料入口。
项目库 / 产品库项目模块入口modules/project/README.md企业微盘项目索引入口。
项目库 / 产品库微盘项目索引modules/project/wedrive-projects/project-index.tsv企业微信微盘项目索引。
项目库 / 产品库公司知识库work_company_knowledge/README.md企业微信微盘派生知识库入口。
项目库 / 产品库公司知识地图work_company_knowledge/knowledge-map.md按角色/主题找公司资料。
项目库 / 产品库在线文档索引work_company_knowledge/online-documents/online-document-index.csv在线文档、表格、接口导入记录。
项目库 / 产品库源文件根目录work_company_knowledge/source-roots.md微盘和本机资料源定位。
项目库 / 产品库平安云厨产品库product_pingan_yunchu/README.md平安云厨方案资料入口。
项目库 / 产品库国信智慧食堂产品库product_guoxin_smart_canteen/README.md国信智慧食堂产品方案入口。
项目库 / 产品库保康智慧食堂产品库product_kezuozhongqi-baokang-middle-school-smart-canteen/README.md保康中学智慧食堂技术材料入口。
项目库 / 产品库食安项目库work_food_safety/README.md食安方案、设备、交付资料。
项目库 / 产品库Store 工作库work_store/architecture/README.mdstore 架构、接口和改动证据入口。
项目库 / 产品库Store 提交前 QA 闸门 Markdownwork_store/qa/store-pre-submit-qa-gate-checklist.mdAgent 提交前回填模板和专项验证矩阵。
项目库 / 产品库AI API 工作库work_ai_api/api-docs/README.mdAI API 接口文档与验证记录。
项目库 / 产品库Android 工作库work_android/README.mdAndroid 终端、设备、SOP 资料入口。
项目库 / 产品库什刹海项目库work_sch1.0new/README.mdsch1.0new 科训项目入口。
项目库 / 产品库青菜项目库work_qc/README.mdqc 项目任务、接口、数据库、工作日志入口。
项目库 / 产品库莱迪森项目库work_laidisen/README.md莱迪森本地部署和交付资料入口。
项目库 / 产品库国信 Windows 部署库work_win_deploy_guoxin/README.md国信部署包、任务和工作日志。
项目库 / 产品库滨州部署文档work_binzhou_health_tech_vocational_college/deployment-records/binzhou-final-deployment-steps-20260707.md滨州健康科技职业学院服务器部署步骤、LNMP、Nginx、数据库初始化和验收入口。
项目库 / 产品库任务总索引control/task-index/tasks.tsv所有任务、证据、复用 skill 的统一索引。
项目库 / 产品库业务仓库清单control/zhctproject-repositories.csvzhctproject 业务仓库登记真源。
项目库 / 产品库本机路径清单control/local-paths.md不同同事本机路径映射。

12. 团队常用方法论工具箱

2026-06-09 从有道云《姜阳的聊天记录_202606081637_来源微信》目标聊天记录图片中恢复了一组产品经理方法论卡片。团队成员不用背所有模型,先按任务场景选工具。

新增:需求四层判断法

2026-06-11 从有道云《一套可复用的【需求四层】判断法,帮你在动手前先看清需求真伪》沉淀。判断需求不要只问“真不真”,先看当前证据停在陈述层、触发层、代价层还是结构层。

需求四层判断法 HTML · 知识 Markdown · PRD 需求分析 skill · 四层判断参考

新增:需求评审 Prompt 作战手册

2026-06-15 从有道云《需求评审总被问懵?这些指令词让你从容应对》沉淀。需求评审前不要只补 PPT,按评审前 10 分钟快速自检、评审前 1 小时系统准备、评审后 30 分钟复盘三段式准备。

需求评审 Prompt HTML · 知识 Markdown · 评审 Prompt 参考 · 需求评审 15 问

新增:PRD 到可验证代码闭环

2026-06-18 从有道云《从 PRD 到可验证代码:AI 需求开发闭环实践》沉淀。从 PRD、接口、设计稿或截图进入代码前,先形成 RequirementSpec、ArchitectureDesign、证据清单和验证策略,再进入实现、架构对齐、编译验证、视觉审计和经验沉淀。

PRD 到可验证代码 HTML · 知识 Markdown · PRD 到代码参考 · Skill + Workflow 规范

场景推荐方法团队成员怎么用
看市场和战略SWOT、PEST、波特五力判断我们强在哪里、弱在哪里、外部机会和威胁是什么,利润被供应商、客户、竞品、替代品还是新进入者挤压。
判断产品是否值得做PMF、MVP、KANO先确认目标客户、未满足需求和最小价值闭环,再区分必须有、期望型和魅力型需求。
判断需求证据深浅需求四层不把“用户感兴趣”直接当成可做需求;先判断证据停在陈述层、触发层、代价层还是结构层,再决定补证、MVP、PRD 或暂缓。
准备需求评审需求评审 Prompt 作战手册评审前 10 分钟快速自检,评审前 1 小时系统准备,评审后 30 分钟复盘;用 AI 模拟开发、测试、设计、业务和领导提前 challenge。
从 PRD 进入代码PRD 到可验证代码闭环先输出 RequirementSpec、ArchitectureDesign、证据清单和验证策略,再进入代码实现、架构对齐、编译验证、视觉审计和经验沉淀。
拆用户体验和行为用户体验地图、福格行为模型、马斯洛需求按角色画完整旅程,再看关键动作是否同时具备动机、能力和触发,并识别底层安全、健康、归属和尊重需求。
排需求和研发优先级RICE、5W2H、PDCA用 Reach x Impact x Confidence / Effort 排优先级,用 5W2H 拆责任、时间、路径和指标,上线后用 PDCA 闭环改进。
做增长和续费AARRR、PMF、用户体验地图管理线索、激活、留存、变现、推荐,避免只看成交,不看复购和转介绍。
做汇报和复盘金字塔原理、STAR汇报先结论后证据;复盘讲清情境、任务、行动和结果。
方法最适合解决的问题
PDCA 循环流程优化、交付问题闭环、功能持续迭代
SWOT 分析组织、产品或项目的内外部态势判断
PEST 分析政策、经济、社会、技术等宏观环境扫描
KANO 模型判断哪些功能是基础项、期望项、惊喜项或无价值项
波特五力判断行业利润结构和竞争压力
PMF判断产品是否真正击中市场刚需
马斯洛需求从底层动机理解用户为什么会买、用、留存
用户体验地图找全流程痛点、爽点、断点和优化机会
金字塔原理写清楚方案、汇报、复盘和领导决策建议
MVP用最小成本验证假设,不一次性堆满功能
福格行为模型设计让用户完成关键动作的动机、能力和触发
AARRR管理获取、激活、留存、变现、推荐增长链路
RICE公平比较需求价值和开发优先级
5W2H把模糊任务拆成可执行计划
STAR结构化表达项目成果、个人贡献和客户案例
需求四层立项前判断需求证据深浅,避免把表态、兴趣、问卷意向误当成真实可开发需求
需求评审 Prompt 作战手册需求评审前准备、角色模拟追问、挑战话术、会议纪要和会后复盘闭环
PRD 到可验证代码闭环把 PRD、接口、设计稿和验证门禁串成可开发、可验证、可回流的 AI 需求开发闭环
请先判断这个问题更适合用哪类方法论:
战略判断、产品验证、用户体验、需求排期、增长续费、汇报复盘。
然后选 1-3 个模型分析,不要把所有模型平均铺开。
真实业务结论必须标证据等级;没有数据时只写假设和待补证据。

13. 产品类 skill 有哪几类,怎么使用

产品类任务不要只说“帮我做个方案”。先判断它属于战略、需求、竞品、方案、专项、接口验证还是视觉表达,再让 Agent 调用对应 skill 组。

产品类 skill 类别典型 skill怎么使用新人示例说法
产品战略 / 经营判断product-development-master-agent, product-strategy-planning, smart-canteen-strategy-agent, digital-business-investor-strategy-lens先锁规划层级、产品线/区域/客户场景和证据等级,再输出产品地图、路线图、经营假设和待补数据。请按产品战略规划和智慧食堂战略 skill,判断这个方向是否值得进入,并输出产品地图、路线图、证据缺口。
需求 / PRD / 系统核验 / 需求评审prd-requirements-analysis, system-grounded-requirement-analysis, wang-huiwen-demand-opportunity, product-material-placement先保存需求原文,再判断真需求/伪需求;涉及立项、付费、SKU 或客户承诺时按需求四层判断证据深浅;需求评审前按三段式准备;从 PRD、接口、设计稿或截图进入代码前,先形成 RequirementSpec、ArchitectureDesign、证据清单和验证策略。请把这段客户需求整理成 PRD,先判断证据层级,再输出可进入开发的规格、架构和验证清单。
市场 / 竞品 / PMF / 上市product-manager-competitive-analysis, wang-huiwen-market-strategy, wang-huiwen-pmf-stp-4p先区分市场体量、竞争格局、第一批客户和最终市场,再推 STP、4P、定价、渠道和推广切入点。请把这些竞品和客户线索拆成竞品结论、PMF 切入点、STP/4P 和下一轮验证动作。
产品设计 / 能力地图 / 方案资产product-design-solution-workspace, product-capability-map-synthesis, food-safety-solution-pdf-agent, product-intro-pdf-generation把零散政策、竞品、客户材料和讨论收敛成能力地图、P0/P1/P2、方案样张、PPT/PDF 或 UI 原型入口。请把这些资料收敛成能力地图和三页产品方案样张,标清证据等级和待确认内容。
智慧食堂 / 食安 / AI 营养专项smart-canteen-agent-native-work-system, smart-canteen-strategy-agent, smart-canteen-showroom-reception, food-safety-solution-pdf-agent围绕智慧营养健康餐厅、食安监管、展厅接访、AI Agent 原生流程,明确人类用户、Agent 用户、读写动作、禁区和人工门禁。请把这个智慧食堂流程设计成 Agent 原生工作系统,并给出展厅/客户演示时的证据和人工门禁。
软件入口 / API / 测试验证api-contract-test-coverage, apifox-api-sync, webapp-testing, post-test-manual-video-delivery当产品需求落到页面、接口、权限、API 契约、测试用例或用户可见功能时,必须同步契约、测试、截图、手册和视频证据。这个功能涉及接口和页面,请同步 Apifox、测试用例、截图和用户手册。
视觉表达 / 人审页面taste-skill-visual-direction, redesign-existing-projects, minimalist-ui, high-end-visual-design, brandkit, image-to-code面向领导、客户、团队培训或人审时,用 HTML/PPT/PDF/视觉参考图提升可读性,但不改业务事实。请按 taste skill 美化这个团队页,保留内容结构,重点提升阅读层级和移动端可读性。

14. 高频项目核心 skill 速查与实例

下面这张表覆盖一批高频项目核心 skill;当前实际 212 个 skill 以目录为准,下一节保留 187 项历史说明快照供检索,PM Skills 详见 PM Skills 团队工作流使用指南。新手不需要背名称,但要会用“场景 + 资料路径 + 输出 + 验收标准”的方式让 Agent 自动路由。

分类skill路径什么时候用实例:可以怎么说
产品战略与经营digital-business-investor-strategy-lensstandards-stack/agent-skills/skills/digital-business-investor-strategy-lens/SKILL.md数科/智慧食堂业务从经营、可复制、持续收入视角分析。请从投资人视角分析这个智慧食堂方向为什么值得做、怎么证明可复制。
产品战略与经营product-development-management-standardstandards-stack/agent-skills/skills/product-development-management-standard/SKILL.md产品开发部管理规范、项目经理检查清单。请按产品开发部管理规范,把这个项目检查清单整理出来。
产品战略与经营product-development-master-agentstandards-stack/agent-skills/skills/product-development-master-agent/SKILL.md产品开发部总蓝图、一总五分和 IPO 路由。请按产品开发部总 Agent 判断这件事归入哪条主线和五模块。
产品战略与经营product-manager-competitive-analysisstandards-stack/agent-skills/skills/product-manager-competitive-analysis/SKILL.md竞品拆解、产品对标和 PM 决策动作。请把这些竞品资料转成产品决策、销售话术和下一轮调研动作。
产品战略与经营product-strategy-planningstandards-stack/agent-skills/skills/product-strategy-planning/SKILL.md产品战略、产品线、区域、渠道、SPAN/APPEALS/PDC/7-2-1。请按产品战略规划方法,把这个产品线做成产品地图和路标。
产品战略与经营smart-canteen-agent-native-work-systemstandards-stack/agent-skills/skills/smart-canteen-agent-native-work-system/SKILL.md把智慧食堂产品或团队流程设计成 Agent 原生工作系统。请把这个智慧食堂流程设计成人和 Agent 共用的工作系统,标出禁区和人工门禁。
产品战略与经营smart-canteen-strategy-agentstandards-stack/agent-skills/skills/smart-canteen-strategy-agent/SKILL.md智慧食堂产品战略、SPAN/PDC/7-2-1 和路线图。请做智慧食堂产品地图和下一阶段路线图,标证据等级。
产品战略与经营wang-huiwen-demand-opportunitystandards-stack/agent-skills/skills/wang-huiwen-demand-opportunity/SKILL.md判断需求/机会是否真实、强烈、可行和有 ROI。请按王慧文需求机会方法判断这个客户需求是不是伪需求。
产品战略与经营wang-huiwen-founder-product-managerstandards-stack/agent-skills/skills/wang-huiwen-founder-product-manager/SKILL.md产品经理从功能交付走向业务结果负责的训练/评审。请按创业产品经理标准评审这个 PM 工作是否只是在交付功能。
产品战略与经营wang-huiwen-market-strategystandards-stack/agent-skills/skills/wang-huiwen-market-strategy/SKILL.md市场体量、规模效应、产业链、频次、存量/增量和入场时机。请按王慧文市场战略方法判断这个市场能不能进入。
产品战略与经营wang-huiwen-pmf-stp-4pstandards-stack/agent-skills/skills/wang-huiwen-pmf-stp-4p/SKILL.mdPMF、第一批客户、STP、4P、定价、渠道和推广。请把这个产品想法拆成 PMF、STP、4P 和第一批客户切入点。
产品需求与方案food-safety-solution-pdf-agentstandards-stack/agent-skills/skills/food-safety-solution-pdf-agent/SKILL.md食安产品方案和 PDF/PPT 资产。请把食安系统材料重构成客户能看的方案样张。
产品需求与方案prd-requirements-analysisstandards-stack/agent-skills/skills/prd-requirements-analysis/SKILL.md需求分析、PRD、需求池筛选、真伪需求判断、需求四层深浅判断,以及从 PRD 到可验证代码的开发前门禁。请把这段想法整理成适合当前复杂度的 PRD,先说明证据层级,再给出可进入开发的规格、架构和验证策略。
产品需求与方案product-capability-map-synthesisstandards-stack/agent-skills/skills/product-capability-map-synthesis/SKILL.md从混乱证据收敛能力地图、P0/P1/P2 和 PRD 拆分。请把这些政策、竞品和客户讨论收敛成能力地图和 P0/P1/P2 范围。
产品需求与方案product-design-solution-workspacestandards-stack/agent-skills/skills/product-design-solution-workspace/SKILL.md产品方案、PRD 支撑、UI 原型、架构图和平安云厨资料。请把这批产品资料归到产品设计工作台,并输出方案/原型入口。
产品需求与方案product-intro-pdf-generationstandards-stack/agent-skills/skills/product-intro-pdf-generation/SKILL.md把产品介绍 Markdown/PPT outline 转成可复用 PDF/PPT。请把这份产品介绍 Markdown 生成适合客户看的 PDF/PPT 结构。
产品需求与方案product-material-placementstandards-stack/agent-skills/skills/product-material-placement/SKILL.md产品资料上传、归档、交接时自动判断落位。请按产品资料落位规则判断这些材料该放到哪个 product/work 目录。
产品需求与方案system-grounded-requirement-analysisstandards-stack/agent-skills/skills/system-grounded-requirement-analysis/SKILL.md从截图/Excel/聊天做需求分析时核验真实 UI、代码、接口和数据。请基于真实系统页面判断这个需求要不要开发,并给 Before/After 方案。
设计与视觉表达brandkitstandards-stack/agent-skills/skills/brandkit/SKILL.md品牌板、Logo 体系、色彩字体和视觉世界。请为这个产品方向生成 brand kit。
设计与视觉表达design-taste-frontendstandards-stack/agent-skills/skills/taste-skill/SKILL.md营销页、作品集、产品介绍页等高质感前端。请按 taste-skill 把这个产品介绍页做成高级但不空洞的页面。
设计与视觉表达design-taste-frontend-v1standards-stack/agent-skills/skills/taste-skill-v1/SKILL.md需要保留 v1 行为兼容时使用。这次沿用旧 taste v1,不要切换新风格。
设计与视觉表达gpt-tastestandards-stack/agent-skills/skills/gpt-tasteskill/SKILL.md高冲击视觉概念和实验性网页方向。请用 gpt-taste 先生成一个视觉方向,但不要直接套到运营后台。
设计与视觉表达high-end-visual-designstandards-stack/agent-skills/skills/soft-skill/SKILL.md客户-facing 高级、柔和、精致视觉。请把这个客户方案页面做得更高级但保持稳重。
设计与视觉表达image-to-codestandards-stack/agent-skills/skills/image-to-code-skill/SKILL.md先有图像参考,再落成 HTML/frontend。请先分析这张参考图,再实现匹配的页面。
设计与视觉表达imagegen-frontend-mobilestandards-stack/agent-skills/skills/imagegen-frontend-mobile/SKILL.md生成移动端 App 屏幕概念图。请生成移动端流程参考图。
设计与视觉表达imagegen-frontend-webstandards-stack/agent-skills/skills/imagegen-frontend-web/SKILL.md生成网站参考图,不直接编码。请先生成 Web 首屏视觉参考图。
设计与视觉表达industrial-brutalist-uistandards-stack/agent-skills/skills/brutalist-skill/SKILL.md工业/终端/海报式实验视觉。这次做一个工业风技术 demo 页面。
设计与视觉表达minimalist-uistandards-stack/agent-skills/skills/minimalist-skill/SKILL.md克制的管理报告、文档和执行页视觉。请用 minimalist 方向优化这份管理报告。
设计与视觉表达redesign-existing-projectsstandards-stack/agent-skills/skills/redesign-skill/SKILL.md审计并升级既有页面,同时保留业务结构。请保留原内容结构,重新设计这个 HTML。
设计与视觉表达stitch-design-tastestandards-stack/agent-skills/skills/stitch-skill/SKILL.mdGoogle Stitch 兼容设计系统和 DESIGN.md。请为这个原型生成 Stitch 兼容的 DESIGN.md。
设计与视觉表达taste-skill-visual-directionstandards-stack/agent-skills/skills/taste-skill-visual-direction/SKILL.mdzhctprompt HTML/Word/PPT/方案材料视觉路由。请用 taste skill 美化这个 HTML,默认走管理报告风格。
研发生命周期debugging-and-error-recoverystandards-stack/agent-skills/skills/debugging-and-error-recovery/SKILL.md测试失败、构建失败或行为异常时系统排查。这个报错请按证据排查根因,记录复现、假设、验证和恢复动作。
研发生命周期documentation-and-adrsstandards-stack/agent-skills/skills/documentation-and-adrs/SKILL.md需要记录设计决策、公共 API 或交付上下文。请把这次架构选择写成 ADR,说明备选方案和取舍。
研发生命周期goalstandards-stack/agent-skills/skills/goal/SKILL.md先对齐目标、成功标准、边界和证据口径。请先用 goal 帮我把这件事的目标、非目标、验收标准和下一步路由写清楚。
研发生命周期incremental-implementationstandards-stack/agent-skills/skills/incremental-implementation/SKILL.md按最小安全切片实施代码或文档改动。请按最小改动先完成第一版,不做无关重构。
研发生命周期planning-and-task-breakdownstandards-stack/agent-skills/skills/planning-and-task-breakdown/SKILL.md把规格拆成可验证的实施切片。请把这个任务拆成 3 到 5 个可验证步骤,并标出每一步证据。
研发生命周期shipping-and-launchstandards-stack/agent-skills/skills/shipping-and-launch/SKILL.md上线、发布、回滚和交付打包前使用。请准备上线检查清单、发布说明和回滚方案。
研发生命周期spec-driven-developmentstandards-stack/agent-skills/skills/spec-driven-development/SKILL.md把自然语言需求整理成可执行规格。请先把这段需求整理成任务规格,明确范围、非目标、验收标准。
研发生命周期test-driven-developmentstandards-stack/agent-skills/skills/test-driven-development/SKILL.md需要测试证明代码或行为变化时使用。请先列测试点,再实现,并运行能覆盖这次改动的验证。
代码质量与安全code-review-and-qualitystandards-stack/agent-skills/skills/code-review-and-quality/SKILL.md代码合并前做正确性、兼容性和风险审查。请按 code-review-and-quality 审这个 diff,优先指出 bug 和缺测试。
代码质量与安全code-simplificationstandards-stack/agent-skills/skills/code-simplification/SKILL.md代码可工作但复杂、重复或难维护时精简。请在行为不变前提下简化这段代码,并说明风险。
代码质量与安全full-output-enforcementstandards-stack/agent-skills/skills/output-skill/SKILL.md防止长文档/代码输出截断、占位和缺段。这个输出很长,请按 full-output-enforcement 分段完整交付。
代码质量与安全karpathy-guidelinesstandards-stack/agent-skills/skills/karpathy-guidelines/SKILL.md编码、审查、重构时保持简洁、精准和可验证。这个代码任务请按 karpathy-guidelines:先澄清,少改动,验证后再说完成。
代码质量与安全security-and-hardeningstandards-stack/agent-skills/skills/security-and-hardening/SKILL.md涉及输入、权限、认证、数据、外部集成时做安全加固。这个接口涉及用户输入,请按 security-and-hardening 检查风险。
云效/Codeup/运行环境aliyun-sls-log-triagestandards-stack/agent-skills/skills/aliyun-sls-log-triage/SKILL.md只读查询阿里云 SLS 日志并脱敏汇报。请根据这个 SLS 链接查订单日志,手机号和请求体要脱敏。
云效/Codeup/运行环境api-contract-test-coveragestandards-stack/agent-skills/skills/api-contract-test-coverage/SKILL.md新增/更新接口时同步契约、测试矩阵和 Apifox/OpenAPI。这个接口变化请同步接口契约、边界测试和 Apifox 场景。
云效/Codeup/运行环境apifox-api-syncstandards-stack/agent-skills/skills/apifox-api-sync/SKILL.md把 OpenAPI/Swagger 增量同步到 Apifox 并核验。请把这些接口同步到 Apifox,先列一级目录让我确认。
云效/Codeup/运行环境async-export-streaming-xlsxstandards-stack/agent-skills/skills/async-export-streaming-xlsx/SKILL.mdstore 异步导出提速但保持 XLSX 下载习惯。消费订单导出太慢,请保持 xlsx,用流式写入优化。
云效/Codeup/运行环境git-workflow-and-commit-governancestandards-stack/agent-skills/skills/git-workflow-and-commit-governance/SKILL.md分支、worktree、提交、cherry-pick、推送和任务号提交治理。请按任务号创建分支和提交,不要混进无关文件。
云效/Codeup/运行环境jsr-db-admin-maintenancestandards-stack/agent-skills/skills/jsr-db-admin-maintenance/SKILL.md金斯瑞 admin 密码更新时间等最小 SQL 维护。请给金斯瑞 admin 密码更新时间的最小 SQL,我自己登录执行。
云效/Codeup/运行环境sxtyj-frontend-dependency-repairstandards-stack/agent-skills/skills/sxtyj-frontend-dependency-repair/SKILL.md修复 sxtyj Vue 前端依赖、node-sass、chromedriver 等构建问题。sxtyj 前端安装失败,请按依赖修复 skill 处理。
云效/Codeup/运行环境yunxiao-batch-subagent-executionstandards-stack/agent-skills/skills/yunxiao-batch-subagent-execution/SKILL.md批量处理多个云效任务,按负责人/状态过滤并隔离 worktree。请筛选某负责人未完成任务,按每个任务独立切片处理。
云效/Codeup/运行环境yunxiao-cli-interactionstandards-stack/agent-skills/skills/yunxiao-cli-interaction/SKILL.md云效工作项读取、创建、更新、评论、工时和状态流转。请用云效 CLI 只读查看这个工作项,并脱敏回读结果。
云效/Codeup/运行环境yunxiao-codeup-pr-associationstandards-stack/agent-skills/skills/yunxiao-codeup-pr-association/SKILL.md云效任务与 Codeup 分支、提交、MR 和 release 流程关联。这个 MR 缺少云效关联,请按 #任务号 规则修正标题和提交证据。
云效/Codeup/运行环境yunxiao-requirement-task-lifecyclestandards-stack/agent-skills/skills/yunxiao-requirement-task-lifecycle/SKILL.md云效需求从拆任务、计划工时、开发、提交到状态更新的生命周期。请拉取这个云效需求,拆开发任务,完成后填实际工时和状态。
云效/Codeup/运行环境yunxiao-smart-canteen-requirementsstandards-stack/agent-skills/skills/yunxiao-smart-canteen-requirements/SKILL.md把截图/聊天/短文本整理成智慧餐厅云效产品类需求。请把这张截图整理成【自研】智慧营养健康餐厅的产品类需求。
云效/Codeup/运行环境yunxiao-wedrive-agentic-collaborationstandards-stack/agent-skills/skills/yunxiao-wedrive-agentic-collaboration/SKILL.md云效、Codeup、微盘和多智能体协作路由。请把这段聊天归一成云效/微盘/代码审查协作流程。
云效/Codeup/运行环境zhct-docker-php-verificationstandards-stack/agent-skills/skills/zhct-docker-php-verification/SKILL.md在 Docker 容器里验证 store/ai_api PHP 语法或 ThinkPHP 命令。请用 zhctapp_v2_0 容器跑 PHP 语法检查,不依赖宿主机 php。
云效/Codeup/运行环境zhct-readme-docker-startstandards-stack/agent-skills/skills/zhct-readme-docker-start/SKILL.md按 zhctprompt/docker/start_zhct.sh 启动本地 Docker,不走旧 compose。请按 README 启动本地智慧餐厅 Docker,并解释 8088 是否需要。
云效/Codeup/运行环境zhctproject-repo-syncstandards-stack/agent-skills/skills/zhctproject-repo-sync/SKILL.md同步 zhctproject 下全部登记业务仓库。请按仓库清单拉取最新代码,遇到本地脏改动就停下报告。
交付/验收/证据delivery-asset-agentstandards-stack/agent-skills/skills/delivery-asset-agent/SKILL.md把策略/能力/证据转成交付资产、售前方案、手册、复盘。请把这些材料整理成可复用交付资产和售前 brief。
交付/验收/证据hardware-adaptation-delivery-gatestandards-stack/agent-skills/skills/hardware-adaptation-delivery-gate/SKILL.md硬件适配和批量部署门禁。这批称重台联调请先列样机、协议和部署风险。
交付/验收/证据operation-manual-videostandards-stack/agent-skills/skills/operation-manual-video/SKILL.md生成正式中文用户操作手册视频。请用操作手册视频流程生成 module.json、截图、配音和 MP4。
交付/验收/证据post-test-manual-video-deliverystandards-stack/agent-skills/skills/post-test-manual-video-delivery/SKILL.md用户可见功能测试通过后补手册、Word、截图和视频材料。功能测试通过后,请自动补用户手册、Word、截图证据和操作视频材料。
交付/验收/证据project-delivery-quantified-acceptancestandards-stack/agent-skills/skills/project-delivery-quantified-acceptance/SKILL.md项目量化验收、防退货、客户签收材料。请把这个项目做成量化验收手册,指标要有样本、公式和待确认项。
交付/验收/证据smart-canteen-showroom-receptionstandards-stack/agent-skills/skills/smart-canteen-showroom-reception/SKILL.md产业园展厅智慧餐厅接访、开机、模型摆放和菜品复位。明天客户参观,请按展厅接访规则列开机和餐台复位清单。
交付/验收/证据software-ip-certification-packagestandards-stack/agent-skills/skills/software-ip-certification-package/SKILL.md软著、CNAS/CB、专利、截图和认证资料包。请准备这个系统的软著资料包,区分草稿和需专业确认的内容。
资料归档与知识库ai-native-knowledge-loopstandards-stack/agent-skills/skills/ai-native-knowledge-loop/SKILL.md会议/评审/上线问题知识回流到 AI Native 模板。请把这次上线问题复盘沉淀成下次可用模板。
资料归档与知识库android-other-knowledge-placementstandards-stack/agent-skills/skills/android-other-knowledge-placement/SKILL.mdAndroid、设备、SOP 等未归属资料需要落位。这份 Android 设备操作流程请放到正确目录,不要新建散乱 work 目录。
资料归档与知识库company-wedrive-knowledge-basestandards-stack/agent-skills/skills/company-wedrive-knowledge-base/SKILL.md企业微信微盘资料、项目交付共享空间和数字技术中心知识库检索。请先查公司微盘知识库,找到这个项目相关资料路径。
资料归档与知识库delivery-recording-knowledge-agentstandards-stack/agent-skills/skills/delivery-recording-knowledge-agent/SKILL.md交付录音、转写、复盘 PDF 和问题闭环沉淀。请把这次交付录音转写成经验包,并关联微盘项目。
资料归档与知识库large-source-asset-managementstandards-stack/agent-skills/skills/large-source-asset-management/SKILL.md大型外部文档、PPT、视频、数据集和微盘资料受控管理。这批大文件不要复制进 Git,请登记来源和派生产物。
资料归档与知识库llm-wiki-knowledge-basestandards-stack/agent-skills/skills/llm-wiki-knowledge-base/SKILL.md把调研、文章、会议、项目资料编译成 Markdown wiki 和可复用 skill。请把这些资料先留 raw 来源索引,再编译成项目 wiki。
资料归档与知识库shichahai-project-knowledgestandards-stack/agent-skills/skills/shichahai-project-knowledge/SKILL.md什刹海体校/科训 sch1.0new 项目知识库。请按什刹海项目知识库回答这个数据库/微盘问题。
资料归档与知识库wechat-article-youdao-intakestandards-stack/agent-skills/skills/wechat-article-youdao-intake/SKILL.md微信公众号文章保存到有道云并回读校验。请把这个公众号链接保存到有道云,若遇验证码就建恢复笔记。
资料归档与知识库youdao-chat-wedrive-project-intakestandards-stack/agent-skills/skills/youdao-chat-wedrive-project-intake/SKILL.md有道/微信群聊笔记快速建立微盘项目文件夹和软硬件表。请把这个群聊笔记转成微盘项目文件夹和软件/硬件跟踪表。
团队治理与沉淀agent-skill-design-qualitystandards-stack/agent-skills/skills/agent-skill-design-quality/SKILL.md新增、维护、评审 skill 时做质量门。请判断这个流程是否真的值得做成 skill,并写正反触发例。
团队治理与沉淀ai-organization-operating-modelstandards-stack/agent-skills/skills/ai-organization-operating-model/SKILL.mdAI 原生组织、平台部、交付组织和项目经营中台设计。请把这个团队协作方式设计成 AI 原生组织 operating model。
团队治理与沉淀ai-prompting-quality-guardrailsstandards-stack/agent-skills/skills/ai-prompting-quality-guardrails/SKILL.md非平凡问答的上下文、证据等级、反方和验证基线。这个问题请先补上下文包、证据等级、反方观点和验证方式。
团队治理与沉淀codex-project-skill-auto-installstandards-stack/agent-skills/skills/codex-project-skill-auto-install/SKILL.md新同事本机自动安装项目 skills 到 ~/.codex/skills。新同事打开 zhctprompt,请自动检查并安装项目 skills。
团队治理与沉淀codex-team-skill-handoffstandards-stack/agent-skills/skills/codex-team-skill-handoff/SKILL.md项目 skills 打包、复制、安装、交接给同事。请把项目 skills 整理成同事可安装的交接说明。
团队治理与沉淀context-handoff-subagent-continuationstandards-stack/agent-skills/skills/context-handoff-subagent-continuation/SKILL.md长任务、上下文接近满或需要 subagent 接续时写回恢复面。这个任务太长,请先写回队列和心跳,再开子线程接续。
团队治理与沉淀continue-executionstandards-stack/agent-skills/skills/continue-execution/SKILL.md上下文压缩、新线程、自动唤醒后从项目状态继续执行。继续上次任务,请先读 HEARTBEAT、QUEUE 和任务索引恢复现场。
团队治理与沉淀controlled-project-workspace-placementstandards-stack/agent-skills/skills/controlled-project-workspace-placement/SKILL.md新增受控项目或纠正项目资料/通用资料落位。这个目录建错了,请按受控项目资料落位规则迁移。
团队治理与沉淀output-format-preferencesstandards-stack/agent-skills/skills/output-format-preferences/SKILL.md选择 Markdown、CSV、HTML 等输出格式。这个资料给 AI 复用写 Markdown,给人看生成 HTML。
团队治理与沉淀task-index-library-maintenancestandards-stack/agent-skills/skills/task-index-library-maintenance/SKILL.md任务、MR、日志排查、证据包进入统一任务索引。这件事完成后请更新 task-index,并重建 tasks.tsv。
团队治理与沉淀task-thread-title-ticket-prefixstandards-stack/agent-skills/skills/task-thread-title-ticket-prefix/SKILL.md新任务线程/subagent 名称加任务号前缀。这个子线程请以 VWCG-xxx 开头命名。
团队治理与沉淀team-document-review-learning-loopstandards-stack/agent-skills/skills/team-document-review-learning-loop/SKILL.md文档/代码修改后形成 Markdown 记录和人审 HTML。这次文档修改请同步写 summary,并生成/更新人审 HTML。
团队治理与沉淀team-weekly-report-generationstandards-stack/agent-skills/skills/team-weekly-report-generation/SKILL.md把零散工作记录整理成数字技术中心周报。请把这些工作记录整理成本周周报,不要编造未提供成果。
团队治理与沉淀thinking-model-answeringstandards-stack/agent-skills/skills/thinking-model-answering/SKILL.md分析、规划、复盘、决策类问题选择合适思维模型。请选 1 到 3 个合适模型分析这个管理问题,并说明证据缺口。
团队治理与沉淀user-taught-skill-capturestandards-stack/agent-skills/skills/user-taught-skill-capture/SKILL.md用户教了以后固定做法时沉淀成项目能力。我刚才教的流程以后都按这个走,请沉淀到项目 skill 或规则入口。
团队治理与沉淀weekly-report-strategic-transparencystandards-stack/agent-skills/skills/weekly-report-strategic-transparency/SKILL.md领导/战略型周报强调主线、决策思路和情报视野。这份周报面向领导,请补主线任务、判断依据和风险。
团队治理与沉淀zhct-root-agent-bootstrapstandards-stack/agent-skills/skills/zhct-root-agent-bootstrap/SKILL.md从 zhct 根目录启动时补齐 zhctprompt 规则加载。当前在 zhct 根目录,请先加载 zhctprompt 的 AGENTS 和 bootstrap。
团队治理与沉淀zhctprompt-online-file-linkstandards-stack/agent-skills/skills/zhctprompt-online-file-link/SKILL.md返回 zhctprompt 文件的线上访问链接。请给我这个 HTML 的线上域名访问链接,并验证资源可用。
团队治理与沉淀zhctprompt-self-update-on-qastandards-stack/agent-skills/skills/zhctprompt-self-update-on-qa/SKILL.md进入 zhct 工作区实质问答前更新控制项目。开始前先检查 zhctprompt 是否能 ff-only pull,脏工作树就报告。
桌面工具与本地环境aimami-installstandards-stack/agent-skills/skills/aimami-install/SKILL.md同事安装 AiMaMi DMG、Gatekeeper 绕过和校验。同事 AiMaMi 打不开,请给 DMG 安装和安全绕过步骤。
桌面工具与本地环境macos-screen-recordingstandards-stack/agent-skills/skills/macos-screen-recording/SKILL.mdmacOS 录屏启动/停止,尤其 ToDesk/桌面 App。请开始录屏保存到桌面,完成后我会让你停止。
验证与证据demo-system-data-screenshot-acceptancestandards-stack/agent-skills/skills/demo-system-data-screenshot-acceptance/SKILL.md演示/近生产数据修改必须有 Before/After、备份和回滚。请先截 Before,再备份、修改演示数据、验证并截 After。
验证与证据dengbao-remediation-evidence-packstandards-stack/agent-skills/skills/dengbao-remediation-evidence-pack/SKILL.md等保整改复测和证据包。请把这次等保整改整理成复测证据包。
验证与证据evidence-classification-agentstandards-stack/agent-skills/skills/evidence-classification-agent/SKILL.md区分 A/B/C/D 证据、事实、推导和假设。请给这些资料标证据等级,哪些能说成事实、哪些只是推测。
验证与证据webapp-testingstandards-stack/agent-skills/skills/webapp-testing/SKILL.md本地 Web 页面需要 Playwright 截图、交互或浏览器证据。请打开这个本地页面,截图并检查移动端是否溢出。

15. 项目 skill 说明快照:历史 187 项,当前实际 212 项

这张长表是此前按 standards-stack/agent-skills/skills/ 生成的 187 项说明快照,保留用于场景检索,不再称为“当前全目录”。2026-07-23 的当前实数是 212;新增、删除或改名后的最终真源始终是实际目录。新手不用背名称,只要按“场景 + 资料路径 + 输出 + 验收标准”描述任务。

统一触发方式:
请先判断这个任务适合哪个项目 skill。
背景:...
资料路径:...
期望输出:...
验收标准:...
如果需要多个 skill,请说明调用顺序和每个 skill 的边界。
分类skill路径什么时候用可以怎么说
团队治理 / Agentagent-skill-design-qualitystandards-stack/agent-skills/skills/agent-skill-design-quality/SKILL.mdLoad when creating, reviewing, refining, or maintaining Codex/Claude/agent skills, especially when deciding whether a workflow deserves a skill, writing skill descriptions, adding gotchas, or checking route spillover.请按 agent-skill-design-quality skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档ai-native-knowledge-loopstandards-stack/agent-skills/skills/ai-native-knowledge-loop/SKILL.mdLoad when meeting recordings, requirement discussions, solution reviews, PR reviews, launch issues, or retrospectives need to be stored as reusable AI Native project knowledge and templates.请按 ai-native-knowledge-loop skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentai-organization-operating-modelstandards-stack/agent-skills/skills/ai-organization-operating-model/SKILL.mdLoad when designing AI-native organization, AI delivery organization, Agent-native R&D, AI platform team roles, project operating cockpit, or Codex/Hermes/OpenClaw/Yunxiao team operating models.请按 ai-organization-operating-model skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentai-prompting-quality-guardrailsstandards-stack/agent-skills/skills/ai-prompting-quality-guardrails/SKILL.mdBottom-layer team Q&A baseline for zhctprompt. Use by default for non-trivial teammate questions, AI task requests, research, writing, reviews, decisions, agent instructions, prompts, and reusable knowledge that need better context, source reliability, anti...请按 ai-prompting-quality-guardrails skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具aimami-installstandards-stack/agent-skills/skills/aimami-install/SKILL.mdUse when a teammate asks how to install AiMaMi on macOS, runs into "无法验证开发者" security prompts, or needs AiMaMi DMG installation help.请按 aimami-install skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIaliyun-sls-log-triagestandards-stack/agent-skills/skills/aliyun-sls-log-triage/SKILL.mdUse when the user provides an Alibaba Cloud SLS console URL or asks to inspect production logs with SLS SDK/CLI, especially for order, print, cron, API, trace-id, queue, and third-party integration troubleshooting.请按 aliyun-sls-log-triage skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档android-other-knowledge-placementstandards-stack/agent-skills/skills/android-other-knowledge-placement/SKILL.mdLoad when the user says a zhctprompt Android-related or uncategorized project artifact should be "沉淀", "记录", "整理成文档", "放到项目里", or "后续都按这个要求", and no specific Android project directory is named. Default durable placement is zhctprompt/work_android/other/.请按 android-other-knowledge-placement skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIapi-contract-test-coveragestandards-stack/agent-skills/skills/api-contract-test-coverage/SKILL.mdUse when creating, updating, importing, or reviewing any API/interface contract. Ensures Apifox/OpenAPI, interface classification, scenario coverage, success/error/boundary tests, performance needs, and test suites are updated together instead of drifting a...请按 api-contract-test-coverage skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIapifox-api-syncstandards-stack/agent-skills/skills/apifox-api-sync/SKILL.mdUse when the user wants to create, update, or sync API documentation into Apifox from OpenAPI/Swagger, especially for the zhct AI运动营养师 project. Handles Apifox token usage, project/folder defaults, backup/export, conservative import, and post-import verifica...请按 apifox-api-sync skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具async-export-streaming-xlsxstandards-stack/agent-skills/skills/async-export-streaming-xlsx/SKILL.mdThis skill should be used when the user asks to optimize async exports, "异步导出流式 XLSX", "消费订单导出提速", "大数据导出", "ExportMealOrder", "ydy_export_log", "不要改 CSV", or when changes to the 9 async-export-backed list pages may affect exported XLSX results.请按 async-export-streaming-xlsx skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉brandkitstandards-stack/agent-skills/skills/brandkit/SKILL.mdPremium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool...请按 brandkit skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉industrial-brutalist-uistandards-stack/agent-skills/skills/brutalist-skill/SKILL.mdRaw mechanical interfaces fusing Swiss typographic print with military terminal aesthetics. Rigid grids, extreme type scale contrast, utilitarian color, analog degradation effects. For data-heavy dashboards, portfolios, or editorial sites that need to feel...请按 industrial-brutalist-ui skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全code-review-and-qualitystandards-stack/agent-skills/skills/code-review-and-quality/SKILL.mdConducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.请按 code-review-and-quality skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具code-simplificationstandards-stack/agent-skills/skills/code-simplification/SKILL.mdSimplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.请按 code-simplification skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentcodex-project-skill-auto-installstandards-stack/agent-skills/skills/codex-project-skill-auto-install/SKILL.mdLoad when a teammate opens zhctprompt in Codex on a local Mac, project skills may not be installed in ~/.codex/skills, or the user asks Codex to automatically install project skills through a child/background thread.请按 codex-project-skill-auto-install skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentcodex-team-skill-handoffstandards-stack/agent-skills/skills/codex-team-skill-handoff/SKILL.mdUse when the user asks to package, copy, install, or share local/project Codex skills for teammates, especially phrases like "复制给其他人使用 Codex", "把本地 skill 沉淀到项目", "Codex skill 交接", or "团队安装 skills".请按 codex-team-skill-handoff skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档company-wedrive-knowledge-basestandards-stack/agent-skills/skills/company-wedrive-knowledge-base/SKILL.mdUse when a task asks about enterprise WeChat WeDrive materials, company knowledge, product/delivery/development shared references, project delivery archives, or the Digital Technology Center knowledge base.请按 company-wedrive-knowledge-base skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentcontext-handoff-subagent-continuationstandards-stack/agent-skills/skills/context-handoff-subagent-continuation/SKILL.mdUse when a zhctprompt task is long-running, the current thread context is close to full, or the user says to open a new thread or subagent to continue work. Preserve queue state, evidence, and recovery instructions before delegating or continuing elsewhere.请按 context-handoff-subagent-continuation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具continue-executionstandards-stack/agent-skills/skills/continue-execution/SKILL.mdResume or continue zhctprompt long-running work from durable project state. Use after context compaction, thread handoff, automation wakeup, or when a new thread/subagent needs to continue an existing queue.请按 continue-execution skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具controlled-project-workspace-placementstandards-stack/agent-skills/skills/controlled-project-workspace-placement/SKILL.mdUse when a new controlled project should be documented inside zhctprompt, or when the user corrects where project-specific versus generic control assets should live.请按 controlled-project-workspace-placement skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全debugging-and-error-recoverystandards-stack/agent-skills/skills/debugging-and-error-recovery/SKILL.mdGuides systematic root-cause debugging. Use when tests fail, builds break, behavior doesn't match expectations, or you encounter any unexpected error. Use when you need a systematic approach to finding and fixing the root cause rather than guessing.请按 debugging-and-error-recovery skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentdelivery-asset-agentstandards-stack/agent-skills/skills/delivery-asset-agent/SKILL.mdThis skill should be used when a task must produce "售前方案", "交付资产", "管理层材料", "样张", "用户手册", "小视频", "演示材料", "PDF 样张", or reusable customer-facing output from existing product evidence.请按 delivery-asset-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档delivery-recording-knowledge-agentstandards-stack/agent-skills/skills/delivery-recording-knowledge-agent/SKILL.mdLoad when交付实施部录音、会议转写、复盘 PDF、交付管理讨论、项目经验包、问题闭环、或“录音存微盘并关联项目”需要沉淀成 zhctprompt 知识库和 agent。请按 delivery-recording-knowledge-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
交付 / 手册 / 验收demo-system-data-screenshot-acceptancestandards-stack/agent-skills/skills/demo-system-data-screenshot-acceptance/SKILL.mdLoad when modifying demo or production-like system data and the user asks for before/after screenshots, backup, rollback, UI acceptance evidence, or a picture-rich comparison list.请按 demo-system-data-screenshot-acceptance skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具dengbao-remediation-evidence-packstandards-stack/agent-skills/skills/dengbao-remediation-evidence-pack/SKILL.mdBuilds a dengbao remediation retest bundle. Use when fixing 等保测评问题, rerunning security regressions, or generating a one-click evidence pack for复测交付.请按 dengbao-remediation-evidence-pack skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具digital-business-investor-strategy-lensstandards-stack/agent-skills/skills/digital-business-investor-strategy-lens/SKILL.mdUse when work involves 数科业务、数字化体育科技、智慧营养健康餐厅、AI 智慧营养健康餐厅, or related weekly reports, strategy, product planning, project reviews, leadership materials, and business logic that must be framed from an investor and operating perspective.请按 digital-business-investor-strategy-lens skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具documentation-and-adrsstandards-stack/agent-skills/skills/documentation-and-adrs/SKILL.mdRecords decisions and documentation. Use when making architectural decisions, changing public APIs, shipping features, or when you need to record context that future engineers and agents will need to understand the codebase.请按 documentation-and-adrs skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentevidence-classification-agentstandards-stack/agent-skills/skills/evidence-classification-agent/SKILL.mdThis skill should be used when sources need to be graded as "A/B/C/D evidence", when the user asks to "区分事实和假设", "证据等级", "证据分类", "资料可信度", or when strategy/PDF outputs must cite evidence boundaries.请按 evidence-classification-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentfood-safety-solution-pdf-agentstandards-stack/agent-skills/skills/food-safety-solution-pdf-agent/SKILL.mdThis skill should be used when the user asks for "康比特食安方案", "食品安全监督管理系统产品简介", "食安 PDF", "方案 PDF", "竞品拆解", "南京小牛", "雄伟科技", "用友餐饮云", "三页样张", or "正式 PDF 重做".请按 food-safety-solution-pdf-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具git-workflow-and-commit-governancestandards-stack/agent-skills/skills/git-workflow-and-commit-governance/SKILL.mdUse when working with Git branches, worktrees, commits, commit message amendments, cherry-picks, pushes, Codeup branches, or VWCG/Yunxiao task-linked code changes; especially when deciding whether to commit, how to format commit messages, how to create task...请按 git-workflow-and-commit-governance skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具goalstandards-stack/agent-skills/skills/goal/SKILL.mdDefine the outcome goal before specification, planning, or implementation. Use when the user invokes `/goal`, asks to clarify the target, success criteria, acceptance boundary, KPI, desired outcome, "目标", "目的", "验收口径", or when a request is too broad and nee...请按 goal skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉gpt-tastestandards-stack/agent-skills/skills/gpt-tasteskill/SKILL.mdElite UX/UI & Advanced GSAP Motion Engineer. Enforces Python-driven true randomization for layout variance, strict AIDA page structure, wide editorial typography (bans 6-line wraps), gapless bento grids, strict GSAP ScrollTriggers (pinning, stacking, scrubb...请按 gpt-taste skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
交付 / 手册 / 验收hardware-adaptation-delivery-gatestandards-stack/agent-skills/skills/hardware-adaptation-delivery-gate/SKILL.mdLoad when smart-canteen delivery involves hardware adaptation, all-in-one devices, weighing tables, SDK/protocol checks, sample devices, deployment milestones, or vendor confirmation.请按 hardware-adaptation-delivery-gate skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉image-to-codestandards-stack/agent-skills/skills/image-to-code-skill/SKILL.mdElite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable,...请按 image-to-code skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉imagegen-frontend-mobilestandards-stack/agent-skills/skills/imagegen-frontend-mobile/SKILL.mdElite mobile app image-generation skill for creating premium, app-native screen concepts and flows. Designed for iOS, Android, and cross-platform mobile products. Prioritizes clean hierarchy, comfortably readable text, strong multi-screen consistency, contr...请按 imagegen-frontend-mobile skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉imagegen-frontend-webstandards-stack/agent-skills/skills/imagegen-frontend-web/SKILL.mdElite frontend image-direction skill for generating premium, conversion-aware website design references. CRITICAL OUTPUT RULE — generate ONE separate horizontal image FOR EVERY section. A landing page with 8 sections produces 8 images. Never compress multip...请按 imagegen-frontend-web skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具incremental-implementationstandards-stack/agent-skills/skills/incremental-implementation/SKILL.mdDelivers changes incrementally. Use when implementing any feature or change that touches more than one file. Use when you're about to write a large amount of code at once, or when a task feels too big to land in one step.请按 incremental-implementation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具jsr-db-admin-maintenancestandards-stack/agent-skills/skills/jsr-db-admin-maintenance/SKILL.mdUse when the user asks for Jinshiyu/Jinsirui (`金斯瑞`, `jsr`) production database admin-account maintenance SQL, especially updating `zhct_jsr.ydy_user.pwd_update_time` for the `admin` account.请按 jsr-db-admin-maintenance skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉karpathy-guidelinesstandards-stack/agent-skills/skills/karpathy-guidelines/SKILL.mdBehavioral engineering guardrails for coding agents. Use when writing, reviewing, or refactoring code to avoid guessing, overengineering, broad unrelated diffs, and unverifiable work.请按 karpathy-guidelines skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具large-source-asset-managementstandards-stack/agent-skills/skills/large-source-asset-management/SKILL.mdManage large external documents, PPTs, videos, datasets, and WeDrive materials as controlled source assets without copying raw binaries into zhctprompt.请按 large-source-asset-management skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档llm-wiki-knowledge-basestandards-stack/agent-skills/skills/llm-wiki-knowledge-base/SKILL.mdThis skill should be used when the user asks to "调研并沉淀知识库", "构建个人 wiki", "用 Obsidian 管理知识库", "把资料自动整理成 wiki", "复杂问答持续更新", "更新项目经验和技能", or mentions Karpathy's LLM Wiki / LLM Knowledge Base method.请按 llm-wiki-knowledge-base skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具macos-screen-recordingstandards-stack/agent-skills/skills/macos-screen-recording/SKILL.mdLoad when the user asks to start, stop, or manage macOS screen recording, especially "录屏", "屏幕录制", "录制一下", "停止录屏", or recording a desktop app such as ToDesk.请按 macos-screen-recording skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉minimalist-uistandards-stack/agent-skills/skills/minimalist-skill/SKILL.mdClean editorial-style interfaces. Warm monochrome palette, typographic contrast, flat bento grids, muted pastels. No gradients, no heavy shadows.请按 minimalist-ui skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
交付 / 手册 / 验收operation-manual-videostandards-stack/agent-skills/skills/operation-manual-video/SKILL.mdLoad when a teammate asks AI to generate, edit, preview, QA, or export a Chinese user operation manual video from source docs, screenshots, product assets, or an existing feature workflow using the codex-video-harness project.请按 operation-manual-video skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具output-format-preferencesstandards-stack/agent-skills/skills/output-format-preferences/SKILL.mdUse when choosing durable storage and user-facing deliverable formats for Jack's zhctprompt work.请按 output-format-preferences skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具full-output-enforcementstandards-stack/agent-skills/skills/output-skill/SKILL.mdOverrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.请按 full-output-enforcement skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具planning-and-task-breakdownstandards-stack/agent-skills/skills/planning-and-task-breakdown/SKILL.mdBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.请按 planning-and-task-breakdown skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / AI 上线检查pm-ai-shipping-intended-vs-implementedstandards-stack/agent-skills/skills/pm-ai-shipping-intended-vs-implemented/SKILL.md"Load when teammates ask for PM Skills pm-ai-shipping/intended-vs-implemented: The method for finding the gap between what a system is supposed to do and what the code actually does — the class of bug generic scanners miss because they have no model of inte...请按 pm-ai-shipping-intended-vs-implemented skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / AI 上线检查pm-ai-shipping-shipping-artifactsstandards-stack/agent-skills/skills/pm-ai-shipping-shipping-artifacts/SKILL.md"Load when teammates ask for PM Skills pm-ai-shipping/shipping-artifacts: The durable documentation set that makes an AI-built (vibe-coded) app reviewable before shipping. A small core every app needs — architecture, user/permission flows, permissions, vari...请按 pm-ai-shipping-shipping-artifacts skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 数据分析pm-data-analytics-ab-test-analysisstandards-stack/agent-skills/skills/pm-data-analytics-ab-test-analysis/SKILL.md"Load when teammates ask for PM Skills pm-data-analytics/ab-test-analysis: Analyze A/B test results with statistical significance, sample size validation, confidence intervals, and ship/extend/stop recommendations. Use when evaluating experiment results, ch...请按 pm-data-analytics-ab-test-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 数据分析pm-data-analytics-cohort-analysisstandards-stack/agent-skills/skills/pm-data-analytics-cohort-analysis/SKILL.md"Load when teammates ask for PM Skills pm-data-analytics/cohort-analysis: Perform cohort analysis on user engagement data — retention curves, feature adoption trends, and segment-level insights. Use when analyzing user retention by cohort, studying feature...请按 pm-data-analytics-cohort-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 数据分析pm-data-analytics-sql-queriesstandards-stack/agent-skills/skills/pm-data-analytics-sql-queries/SKILL.md"Load when teammates ask for PM Skills pm-data-analytics/sql-queries: Generate SQL queries from natural language descriptions. Supports BigQuery, PostgreSQL, MySQL, and other dialects. Reads database schemas from uploaded diagrams or documentation. Use when...请按 pm-data-analytics-sql-queries skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-brainstorm-okrsstandards-stack/agent-skills/skills/pm-execution-brainstorm-okrs/SKILL.md"Load when teammates ask for PM Skills pm-execution/brainstorm-okrs: Brainstorm team-level OKRs aligned with company objectives — qualitative objectives with measurable key results. Use when setting quarterly OKRs, aligning team goals with company strategy,...请按 pm-execution-brainstorm-okrs skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-create-prdstandards-stack/agent-skills/skills/pm-execution-create-prd/SKILL.md"Load when teammates ask for PM Skills pm-execution/create-prd: Create a Product Requirements Document using a comprehensive 8-section template covering problem, objectives, segments, value propositions, solution, and release planning. Use when writing a PR...请按 pm-execution-create-prd skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-dummy-datasetstandards-stack/agent-skills/skills/pm-execution-dummy-dataset/SKILL.md"Load when teammates ask for PM Skills pm-execution/dummy-dataset: Generate realistic dummy datasets for testing with customizable columns, constraints, and output formats (CSV, JSON, SQL, Python script). Use when creating test data, building mock datasets,...请按 pm-execution-dummy-dataset skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-job-storiesstandards-stack/agent-skills/skills/pm-execution-job-stories/SKILL.md"Load when teammates ask for PM Skills pm-execution/job-stories: Create job stories using the 'When [situation], I want to [motivation], so I can [outcome]' format with detailed acceptance criteria. Use when writing job stories, creating JTBD-style backlog...请按 pm-execution-job-stories skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-outcome-roadmapstandards-stack/agent-skills/skills/pm-execution-outcome-roadmap/SKILL.md"Load when teammates ask for PM Skills pm-execution/outcome-roadmap: Transform an output-focused roadmap into an outcome-focused one that communicates strategic intent. Rewrites initiatives as outcome statements reflecting user and business impacts. Use whe...请按 pm-execution-outcome-roadmap skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-pre-mortemstandards-stack/agent-skills/skills/pm-execution-pre-mortem/SKILL.md"Load when teammates ask for PM Skills pm-execution/pre-mortem: Run a pre-mortem risk analysis on a PRD or launch plan. Categorizes risks as Tigers (real problems), Paper Tigers (overblown concerns), and Elephants (unspoken worries), then classifies as laun...请按 pm-execution-pre-mortem skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-prioritization-frameworksstandards-stack/agent-skills/skills/pm-execution-prioritization-frameworks/SKILL.md"Load when teammates ask for PM Skills pm-execution/prioritization-frameworks: Reference guide to 9 prioritization frameworks with formulas, when-to-use guidance, and templates — RICE, ICE, Kano, MoSCoW, Opportunity Score, and more. Use when selecting a pri...请按 pm-execution-prioritization-frameworks skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-release-notesstandards-stack/agent-skills/skills/pm-execution-release-notes/SKILL.md"Load when teammates ask for PM Skills pm-execution/release-notes: Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release no...请按 pm-execution-release-notes skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-retrostandards-stack/agent-skills/skills/pm-execution-retro/SKILL.md"Load when teammates ask for PM Skills pm-execution/retro: Facilitate a structured sprint retrospective — what went well, what didn't, and prioritized action items with owners and deadlines. Use when running a retrospective, reflecting on a sprint, creating...请按 pm-execution-retro skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-sprint-planstandards-stack/agent-skills/skills/pm-execution-sprint-plan/SKILL.md"Load when teammates ask for PM Skills pm-execution/sprint-plan: Plan a sprint with capacity estimation, story selection, dependency mapping, and risk identification. Use when preparing for sprint planning, estimating team capacity, selecting stories, or ba...请按 pm-execution-sprint-plan skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-stakeholder-mapstandards-stack/agent-skills/skills/pm-execution-stakeholder-map/SKILL.md"Load when teammates ask for PM Skills pm-execution/stakeholder-map: Build a stakeholder map using a power/interest grid, identify communication strategies per quadrant, and generate a communication plan. Use when managing stakeholders, preparing for a laun...请按 pm-execution-stakeholder-map skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-strategy-red-teamstandards-stack/agent-skills/skills/pm-execution-strategy-red-team/SKILL.md"Load when teammates ask for PM Skills pm-execution/strategy-red-team: Red-team a PRD, roadmap, or strategy by attacking its load-bearing assumptions before reality does. Steelmans then attacks each claim, ranks failure modes by impact × likelihood × cheapn...请按 pm-execution-strategy-red-team skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-summarize-meetingstandards-stack/agent-skills/skills/pm-execution-summarize-meeting/SKILL.md"Load when teammates ask for PM Skills pm-execution/summarize-meeting: Summarize a meeting transcript into structured notes with date, participants, topic, key decisions, summary points, and action items. Use when processing meeting recordings, creating mee...请按 pm-execution-summarize-meeting skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-test-scenariosstandards-stack/agent-skills/skills/pm-execution-test-scenarios/SKILL.md"Load when teammates ask for PM Skills pm-execution/test-scenarios: Create comprehensive test scenarios from user stories with test objectives, starting conditions, user roles, step-by-step actions, and expected outcomes. Use when writing QA test cases, cre...请按 pm-execution-test-scenarios skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-user-storiesstandards-stack/agent-skills/skills/pm-execution-user-stories/SKILL.md"Load when teammates ask for PM Skills pm-execution/user-stories: Create user stories following the 3 C's (Card, Conversation, Confirmation) and INVEST criteria with descriptions, design links, and acceptance criteria. Use when writing user stories, breakin...请按 pm-execution-user-stories skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 执行交付pm-execution-wwasstandards-stack/agent-skills/skills/pm-execution-wwas/SKILL.md"Load when teammates ask for PM Skills pm-execution/wwas: Create product backlog items in Why-What-Acceptance format — independent, valuable, testable items with strategic context. Use when writing structured backlog items, breaking features into work items...请按 pm-execution-wwas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-beachhead-segmentstandards-stack/agent-skills/skills/pm-go-to-market-beachhead-segment/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/beachhead-segment: Identify the first beachhead market segment for a product launch. Evaluates segments against burning pain, willingness to pay, winnable market share, and referral potential. Use when...请按 pm-go-to-market-beachhead-segment skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-competitive-battlecardstandards-stack/agent-skills/skills/pm-go-to-market-competitive-battlecard/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/competitive-battlecard: Create sales-ready competitive battlecards comparing your product against a specific competitor — positioning, feature comparison, objection handling, and win/loss patterns. Use...请按 pm-go-to-market-competitive-battlecard skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-growth-loopsstandards-stack/agent-skills/skills/pm-go-to-market-growth-loops/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/growth-loops: Identify growth loops (flywheels) for sustainable traction. Evaluates 5 loop types: Viral, Usage, Collaboration, User-Generated, and Referral. Use when designing growth mechanisms, buildin...请按 pm-go-to-market-growth-loops skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-gtm-motionsstandards-stack/agent-skills/skills/pm-go-to-market-gtm-motions/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/gtm-motions: Identify the best GTM motions and tools across 7 motion types: Inbound, Outbound, Paid Digital, Community, Partners, ABM, and PLG. Use when selecting marketing channels, choosing between in...请按 pm-go-to-market-gtm-motions skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-gtm-strategystandards-stack/agent-skills/skills/pm-go-to-market-gtm-strategy/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/gtm-strategy: Create a go-to-market strategy covering marketing channels, messaging, success metrics, and launch timeline. Use when planning a product launch, creating a GTM plan from scratch, or defini...请按 pm-go-to-market-gtm-strategy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / GTMpm-go-to-market-ideal-customer-profilestandards-stack/agent-skills/skills/pm-go-to-market-ideal-customer-profile/SKILL.md"Load when teammates ask for PM Skills pm-go-to-market/ideal-customer-profile: Identify the Ideal Customer Profile (ICP) from research data with demographics, behaviors, JTBD, and needs. Use when defining your ICP, analyzing PMF survey data, or understandin...请按 pm-go-to-market-ideal-customer-profile skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-competitor-analysisstandards-stack/agent-skills/skills/pm-market-research-competitor-analysis/SKILL.md"Load when teammates ask for PM Skills pm-market-research/competitor-analysis: Analyze competitors with strengths, weaknesses, and differentiation opportunities. Identifies direct competitors and maps the competitive landscape. Use when doing competitive re...请按 pm-market-research-competitor-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-customer-journey-mapstandards-stack/agent-skills/skills/pm-market-research-customer-journey-map/SKILL.md"Load when teammates ask for PM Skills pm-market-research/customer-journey-map: Create an end-to-end customer journey map with stages, touchpoints, emotions, pain points, and opportunities. Use when mapping the customer experience, identifying friction poin...请按 pm-market-research-customer-journey-map skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-market-segmentsstandards-stack/agent-skills/skills/pm-market-research-market-segments/SKILL.md"Load when teammates ask for PM Skills pm-market-research/market-segments: Identify 3-5 potential customer segments with demographics, JTBD, and product fit analysis. Use when exploring market segments, identifying target audiences, evaluating new markets,...请按 pm-market-research-market-segments skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-market-sizingstandards-stack/agent-skills/skills/pm-market-research-market-sizing/SKILL.md"Load when teammates ask for PM Skills pm-market-research/market-sizing: Estimate market size using TAM, SAM, and SOM with top-down and bottom-up approaches. Use when sizing a market opportunity, estimating addressable market, preparing for investor pitches...请按 pm-market-research-market-sizing skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-sentiment-analysisstandards-stack/agent-skills/skills/pm-market-research-sentiment-analysis/SKILL.md"Load when teammates ask for PM Skills pm-market-research/sentiment-analysis: Analyze user feedback data to identify segments with sentiment scores, JTBD, and product satisfaction insights. Use when analyzing user feedback at scale, running sentiment analys...请按 pm-market-research-sentiment-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-user-personasstandards-stack/agent-skills/skills/pm-market-research-user-personas/SKILL.md"Load when teammates ask for PM Skills pm-market-research/user-personas: Create refined user personas from research data — 3 personas with JTBD, pains, gains, and unexpected insights. Use when building personas from survey data, creating user profiles from...请按 pm-market-research-user-personas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 市场研究pm-market-research-user-segmentationstandards-stack/agent-skills/skills/pm-market-research-user-segmentation/SKILL.md"Load when teammates ask for PM Skills pm-market-research/user-segmentation: Segment users from feedback data based on behavior, JTBD, and needs. Identifies at least 3 distinct user segments. Use when segmenting a user base, analyzing diverse user feedback,...请按 pm-market-research-user-segmentation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 增长营销pm-marketing-growth-marketing-ideasstandards-stack/agent-skills/skills/pm-marketing-growth-marketing-ideas/SKILL.md"Load when teammates ask for PM Skills pm-marketing-growth/marketing-ideas: Generate 5 creative, cost-effective marketing ideas with channels, messaging, and engagement rationale. Use when brainstorming marketing campaigns, planning product promotion, or lo...请按 pm-marketing-growth-marketing-ideas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 增长营销pm-marketing-growth-north-star-metricstandards-stack/agent-skills/skills/pm-marketing-growth-north-star-metric/SKILL.md"Load when teammates ask for PM Skills pm-marketing-growth/north-star-metric: Define a North Star Metric and 3-5 supporting input metrics that form a metrics constellation. Classify the business game (Attention, Transaction, Productivity) and validate again...请按 pm-marketing-growth-north-star-metric skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 增长营销pm-marketing-growth-positioning-ideasstandards-stack/agent-skills/skills/pm-marketing-growth-positioning-ideas/SKILL.md"Load when teammates ask for PM Skills pm-marketing-growth/positioning-ideas: Brainstorm product positioning ideas differentiated from competitors. Identifies top competitors and generates positioning statements with rationale. Use when developing product p...请按 pm-marketing-growth-positioning-ideas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 增长营销pm-marketing-growth-product-namestandards-stack/agent-skills/skills/pm-marketing-growth-product-name/SKILL.md"Load when teammates ask for PM Skills pm-marketing-growth/product-name: Brainstorm 5 unique, memorable product names with rationale aligned to brand values and target audience. Use when naming a new product, rebranding, or exploring product name ideas."请按 pm-marketing-growth-product-name skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 增长营销pm-marketing-growth-value-prop-statementsstandards-stack/agent-skills/skills/pm-marketing-growth-value-prop-statements/SKILL.md"Load when teammates ask for PM Skills pm-marketing-growth/value-prop-statements: Generate value proposition statements for marketing, sales, and onboarding from existing value propositions. Use when writing marketing copy, creating sales messaging, or craf...请按 pm-marketing-growth-value-prop-statements skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-analyze-feature-requestsstandards-stack/agent-skills/skills/pm-product-discovery-analyze-feature-requests/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/analyze-feature-requests: Analyze and prioritize a list of feature requests by theme, strategic alignment, impact, effort, and risk. Use when reviewing customer feature requests, triaging a backlog...请按 pm-product-discovery-analyze-feature-requests skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-brainstorm-experiments-existingstandards-stack/agent-skills/skills/pm-product-discovery-brainstorm-experiments-existing/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/brainstorm-experiments-existing: Design experiments to test assumptions for an existing product — prototypes, A/B tests, spikes, and other low-effort validation methods. Use when validating assumpt...请按 pm-product-discovery-brainstorm-experiments-existing skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-brainstorm-experiments-newstandards-stack/agent-skills/skills/pm-product-discovery-brainstorm-experiments-new/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/brainstorm-experiments-new: Design lean startup experiments (pretotypes) for a new product. Creates XYZ hypotheses and suggests low-effort validation methods like landing pages, explainer videos, a...请按 pm-product-discovery-brainstorm-experiments-new skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-brainstorm-ideas-existingstandards-stack/agent-skills/skills/pm-product-discovery-brainstorm-ideas-existing/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/brainstorm-ideas-existing: Brainstorm product ideas for an existing product using multi-perspective ideation from PM, Designer, and Engineer viewpoints. Use when generating new feature ideas, brain...请按 pm-product-discovery-brainstorm-ideas-existing skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-brainstorm-ideas-newstandards-stack/agent-skills/skills/pm-product-discovery-brainstorm-ideas-new/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/brainstorm-ideas-new: Brainstorm feature ideas for a new product in initial discovery from PM, Designer, and Engineer perspectives. Use when starting product discovery for a new product, exploring...请按 pm-product-discovery-brainstorm-ideas-new skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-identify-assumptions-existingstandards-stack/agent-skills/skills/pm-product-discovery-identify-assumptions-existing/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/identify-assumptions-existing: Identify risky assumptions for a feature idea in an existing product across Value, Usability, Viability, and Feasibility. Uses multi-perspective devil's advocate thin...请按 pm-product-discovery-identify-assumptions-existing skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-identify-assumptions-newstandards-stack/agent-skills/skills/pm-product-discovery-identify-assumptions-new/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/identify-assumptions-new: Identify risky assumptions for a new product idea across 8 risk categories including Go-to-Market, Strategy, and Team. Use when evaluating startup risks, assessing a new p...请按 pm-product-discovery-identify-assumptions-new skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-interview-scriptstandards-stack/agent-skills/skills/pm-product-discovery-interview-script/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/interview-script: Create a structured customer interview script with JTBD probing questions, warm-up, core exploration, and wrap-up sections. Follows The Mom Test principles — no leading questions,...请按 pm-product-discovery-interview-script skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-metrics-dashboardstandards-stack/agent-skills/skills/pm-product-discovery-metrics-dashboard/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/metrics-dashboard: Define and design a product metrics dashboard with key metrics, data sources, visualization types, and alert thresholds. Use when creating a metrics dashboard, defining KPIs, set...请按 pm-product-discovery-metrics-dashboard skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-opportunity-solution-treestandards-stack/agent-skills/skills/pm-product-discovery-opportunity-solution-tree/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/opportunity-solution-tree: Build an Opportunity Solution Tree (OST) to structure product discovery — map a desired outcome to opportunities, solutions, and experiments. Based on Teresa Torres' Cont...请按 pm-product-discovery-opportunity-solution-tree skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-prioritize-assumptionsstandards-stack/agent-skills/skills/pm-product-discovery-prioritize-assumptions/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/prioritize-assumptions: Prioritize assumptions using an Impact × Risk matrix and suggest experiments for each. Use when triaging a list of assumptions, deciding what to test first, or applying the...请按 pm-product-discovery-prioritize-assumptions skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-prioritize-featuresstandards-stack/agent-skills/skills/pm-product-discovery-prioritize-features/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/prioritize-features: Prioritize a backlog of feature ideas based on impact, effort, risk, and strategic alignment with top 5 recommendations. Use when prioritizing a feature backlog, making scope d...请按 pm-product-discovery-prioritize-features skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品发现pm-product-discovery-summarize-interviewstandards-stack/agent-skills/skills/pm-product-discovery-summarize-interview/SKILL.md"Load when teammates ask for PM Skills pm-product-discovery/summarize-interview: Summarize a customer interview transcript into a structured template with JTBD, satisfaction signals, and action items. Use when processing interview recordings or transcripts,...请按 pm-product-discovery-summarize-interview skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-ansoff-matrixstandards-stack/agent-skills/skills/pm-product-strategy-ansoff-matrix/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/ansoff-matrix: Generate an Ansoff Matrix analysis mapping growth strategies across market penetration, market development, product development, and diversification. Use when considering growth optio...请按 pm-product-strategy-ansoff-matrix skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-business-modelstandards-stack/agent-skills/skills/pm-product-strategy-business-model/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/business-model: Generate a Business Model Canvas with all 9 building blocks. Use when creating a business model, documenting how a business creates value, or analyzing an existing business model."请按 pm-product-strategy-business-model skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-lean-canvasstandards-stack/agent-skills/skills/pm-product-strategy-lean-canvas/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/lean-canvas: Generate a Lean Canvas with problem, solution, metrics, cost structure, UVP, unfair advantage, channels, segments, and revenue. Use when exploring a lean startup canvas, testing a busin...请按 pm-product-strategy-lean-canvas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-monetization-strategystandards-stack/agent-skills/skills/pm-product-strategy-monetization-strategy/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/monetization-strategy: Brainstorm 3-5 monetization strategies with audience fit, risks, and validation experiments. Use when exploring revenue models, evaluating pricing strategies, or deciding how...请按 pm-product-strategy-monetization-strategy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-pestle-analysisstandards-stack/agent-skills/skills/pm-product-strategy-pestle-analysis/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/pestle-analysis: Perform a PESTLE analysis covering Political, Economic, Social, Technological, Legal, and Environmental factors. Use when assessing the macro environment, doing strategic planning,...请按 pm-product-strategy-pestle-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-porters-five-forcesstandards-stack/agent-skills/skills/pm-product-strategy-porters-five-forces/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/porters-five-forces: Perform Porter's Five Forces analysis — competitive rivalry, supplier power, buyer power, threat of substitutes, and threat of new entrants. Use when analyzing industry dynamics...请按 pm-product-strategy-porters-five-forces skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-pricing-strategystandards-stack/agent-skills/skills/pm-product-strategy-pricing-strategy/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/pricing-strategy: Analyze and design pricing strategies including pricing models, competitive pricing analysis, willingness-to-pay estimation, and price elasticity. Use when setting prices, evaluati...请按 pm-product-strategy-pricing-strategy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-product-strategystandards-stack/agent-skills/skills/pm-product-strategy-product-strategy/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/product-strategy: Create a comprehensive product strategy using the 9-section Product Strategy Canvas — vision, segments, costs, value propositions, trade-offs, metrics, growth, capabilities, and de...请按 pm-product-strategy-product-strategy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-product-visionstandards-stack/agent-skills/skills/pm-product-strategy-product-vision/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/product-vision: Brainstorm an inspiring, achievable, and emotional product vision that motivates teams and aligns stakeholders. Use when defining or refining a product vision, creating a vision stat...请按 pm-product-strategy-product-vision skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-startup-canvasstandards-stack/agent-skills/skills/pm-product-strategy-startup-canvas/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/startup-canvas: Generate a Startup Canvas combining Product Strategy (9 sections) and Business Model (costs + revenue) for a new product. An alternative to BMC and Lean Canvas that separates strateg...请按 pm-product-strategy-startup-canvas skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-swot-analysisstandards-stack/agent-skills/skills/pm-product-strategy-swot-analysis/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/swot-analysis: Perform a detailed SWOT analysis — strengths, weaknesses, opportunities, and threats with actionable recommendations. Use when doing strategic assessment, competitive analysis, or eva...请按 pm-product-strategy-swot-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 产品战略pm-product-strategy-value-propositionstandards-stack/agent-skills/skills/pm-product-strategy-value-proposition/SKILL.md"Load when teammates ask for PM Skills pm-product-strategy/value-proposition: Design a detailed value proposition using a 6-part JTBD template — Who, Why, What before, How, What after, Alternatives. Use when creating a value proposition, analyzing customer...请按 pm-product-strategy-value-proposition skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentpm-skills-team-workflow-routerstandards-stack/agent-skills/skills/pm-skills-team-workflow-router/SKILL.md"Load when teammates ask to use the imported PM Skills marketplace for complete PM workflows such as quick PRD, strategy canvas, user research, competitive analysis, growth/GTM, data analysis, or AI shipping checks."请按 pm-skills-team-workflow-router skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 工具箱pm-toolkit-draft-ndastandards-stack/agent-skills/skills/pm-toolkit-draft-nda/SKILL.md"Load when teammates ask for PM Skills pm-toolkit/draft-nda: Draft a detailed Non-Disclosure Agreement between two parties covering information types, jurisdiction, and clauses needing legal review. Use when creating confidentiality agreements or preparing...请按 pm-toolkit-draft-nda skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 工具箱pm-toolkit-grammar-checkstandards-stack/agent-skills/skills/pm-toolkit-grammar-check/SKILL.md"Load when teammates ask for PM Skills pm-toolkit/grammar-check: Identify grammar, logical, and flow errors in text and suggest targeted fixes without rewriting the entire text. Use when proofreading content, checking writing quality, or reviewing a draft."请按 pm-toolkit-grammar-check skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 工具箱pm-toolkit-privacy-policystandards-stack/agent-skills/skills/pm-toolkit-privacy-policy/SKILL.md"Load when teammates ask for PM Skills pm-toolkit/privacy-policy: Draft a detailed privacy policy covering data types, jurisdiction, GDPR and compliance considerations, and clauses needing legal review. Use when creating a privacy policy, updating data prot...请按 pm-toolkit-privacy-policy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
PM Skills / 工具箱pm-toolkit-review-resumestandards-stack/agent-skills/skills/pm-toolkit-review-resume/SKILL.md"Load when teammates ask for PM Skills pm-toolkit/review-resume: Comprehensive PM resume review and tailoring against 10 best practices including XYZ+S formula, keyword optimization, job-specific tailoring, and structure. Use when reviewing a PM resume, pre...请按 pm-toolkit-review-resume skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全post-test-manual-video-deliverystandards-stack/agent-skills/skills/post-test-manual-video-delivery/SKILL.mdUse after a user-facing feature or bugfix has passed functional testing. Automatically generate or update the user manual in Markdown and Word format, collect screenshot evidence, and prepare or record a short operation-guide video when the environment is s...请按 post-test-manual-video-delivery skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂prd-requirements-analysisstandards-stack/agent-skills/skills/prd-requirements-analysis/SKILL.mdUse when Codex needs to analyze requirements, identify real vs. false requirements, filter or prioritize a requirement pool/backlog, write or review a PRD, run pre-review checks before a requirement review meeting, find the 15 most important review question...请按 prd-requirements-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-buying-points-translationstandards-stack/agent-skills/skills/product-buying-points-translation/SKILL.mdLoad when turning product selling points, features, device specs, product packages, or方案话术 into customer buying points, especially when the user says 卖点, 买点, 产品卖不动, 客户为什么买, 为什么选我们, or 销售话术.请按 product-buying-points-translation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-capability-map-synthesisstandards-stack/agent-skills/skills/product-capability-map-synthesis/SKILL.mdUse when a product design task needs to quickly converge from messy evidence, policies, competitor/project materials, and stakeholder discussion into a review-ready product capability map, P0/P1/P2 scope, module/page translation, or XMind capability artifact.请按 product-capability-map-synthesis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-design-solution-workspacestandards-stack/agent-skills/skills/product-design-solution-workspace/SKILL.mdUse when zhctprompt work involves product design, solution proposals, UI prototypes, design specs, architecture diagrams, or Ping An Yun Chu materials.请按 product-design-solution-workspace skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-development-management-standardstandards-stack/agent-skills/skills/product-development-management-standard/SKILL.mdUse when a task involves 产品开发部工作管理规范, 项目管理规范, 云效 project 管理, 项目经理执行检查清单, or copying these management standards to teammates.请按 product-development-management-standard skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-development-master-agentstandards-stack/agent-skills/skills/product-development-master-agent/SKILL.mdThis skill should be used when the user asks to "把产品开发部规划蓝图放到项目内", "构建整个项目蓝图", "这个作为总的 agent", "软件硬件 AI 工程化蓝图", "把工作填进蓝图", "按 IPO 使用项目", "Prompt OS", "三条主线", or asks how product-development tasks across smart-canteen strategy, food-safety solution assets, s...请按 product-development-master-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-intro-pdf-generationstandards-stack/agent-skills/skills/product-intro-pdf-generation/SKILL.mdUse when a Markdown product introduction, PPT outline, competitor brief, or video workflow must become a reusable, polished product-introduction PDF/PPT asset for 康比特 or zhctprompt product work.请按 product-intro-pdf-generation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-manager-competitive-analysisstandards-stack/agent-skills/skills/product-manager-competitive-analysis/SKILL.mdLoad when the user asks for 产品经理竞品分析, 竞品分析, 竞对分析, 竞品拆解, 产品对标, or turning Youdao/Obsidian competitor notes into PM knowledge, templates, or product decisions.请按 product-manager-competitive-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-material-placementstandards-stack/agent-skills/skills/product-material-placement/SKILL.mdLoad when product people generate, upload, hand off, archive, or ask Codex to organize product materials so the materials are automatically placed under the correct zhctprompt path.请按 product-material-placement skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂product-strategy-planningstandards-stack/agent-skills/skills/product-strategy-planning/SKILL.mdUse when the user asks for 产品战略规划, 产品规划, 产品线战略规划, 公司级产品战略, 区域产品规划, SPAN, PDC, APPEALS, 7-2-1, 产品地图, 产品组合, 产品路标, 渠道规划, or 战略绩效看板.请按 product-strategy-planning skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
交付 / 手册 / 验收project-delivery-quantified-acceptancestandards-stack/agent-skills/skills/project-delivery-quantified-acceptance/SKILL.mdLoad when a project needs quantified acceptance, customer signoff, anti-return delivery documents, accuracy metrics, learning-cycle wording, or sales-facing confirmation scripts.请按 project-delivery-quantified-acceptance skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉redesign-existing-projectsstandards-stack/agent-skills/skills/redesign-skill/SKILL.mdUpgrades existing websites and apps to premium quality. Audits current design, identifies generic AI patterns, and applies high-end design standards without breaking functionality. Works with any CSS framework or vanilla CSS.请按 redesign-existing-projects skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全security-and-hardeningstandards-stack/agent-skills/skills/security-and-hardening/SKILL.mdHardens code against vulnerabilities. Use when handling user input, authentication, data storage, or external integrations. Use when building any feature that accepts untrusted data, manages user sessions, or interacts with third-party services.请按 security-and-hardening skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档shichahai-project-knowledgestandards-stack/agent-skills/skills/shichahai-project-knowledge/SKILL.mdUse when the user asks about 什刹海体校、什刹海科训、科学训练数字化平台、sch1.0new、侯爽、交钥匙、目录链、数据库账号密码、企业微信群聊或企业微盘材料。请按 shichahai-project-knowledge skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具shipping-and-launchstandards-stack/agent-skills/skills/shipping-and-launch/SKILL.mdPrepares production launches. Use when preparing to deploy to production. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.请按 shipping-and-launch skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂smart-canteen-agent-native-work-systemstandards-stack/agent-skills/skills/smart-canteen-agent-native-work-system/SKILL.mdLoad when the user asks to make smart-canteen products or team workflows "AI Agent 必备能力", "Agent 原生", "公司级 Agent", "让 Codex/Claude 操作业务系统", "人和 Agent 一起工作", or to turn PM/design/delivery work into agent-readable executable evidence.请按 smart-canteen-agent-native-work-system skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂smart-canteen-sales-forecastingstandards-stack/agent-skills/skills/smart-canteen-sales-forecasting/SKILL.mdLoad when the user asks for 智慧食堂销售预测, 销售漏斗概率化, 商机评分, 线索评分, 赢单概率, 月度签约预测, B2B销售预测AI化, or turning smart-canteen presales data into forecastable pipeline metrics.请按 smart-canteen-sales-forecasting skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂smart-canteen-showroom-receptionstandards-stack/agent-skills/skills/smart-canteen-showroom-reception/SKILL.mdUse when the task involves产业园展厅、智慧餐厅接访、展厅开机、默认餐台菜品、模型摆放、布菲炉复位或客户参观演示准备.请按 smart-canteen-showroom-reception skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂smart-canteen-strategy-agentstandards-stack/agent-skills/skills/smart-canteen-strategy-agent/SKILL.mdThis skill should be used when the user asks for "智慧食堂产品战略", "职工营养健康体系", "智慧食堂战略规划 V0.1", "产品地图", "SPAN", "PDC", "7-2-1", "产品路标", "区域渠道规划", or "战略绩效看板".请按 smart-canteen-strategy-agent skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉high-end-visual-designstandards-stack/agent-skills/skills/soft-skill/SKILL.mdTeaches the AI to design like a high-end agency. Defines the exact fonts, spacing, shadows, card structures, and animations that make a website feel expensive. Blocks all the common defaults that make AI designs look cheap or generic.请按 high-end-visual-design skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具software-ip-certification-packagestandards-stack/agent-skills/skills/software-ip-certification-package/SKILL.mdLoad when preparing software copyright, CNAS/CB certification, patent-mining, screenshot lists, version descriptions, or weekly IP/certification material packages.请按 software-ip-certification-package skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具spec-driven-developmentstandards-stack/agent-skills/skills/spec-driven-development/SKILL.mdCreates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when requirements are unclear, ambiguous, or only exist as a vague idea.请按 spec-driven-development skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉stitch-design-tastestandards-stack/agent-skills/skills/stitch-skill/SKILL.mdSemantic Design System Skill for Google Stitch. Generates agent-friendly DESIGN.md files that enforce premium, anti-generic UI standards — strict typography, calibrated color, asymmetric layouts, perpetual micro-motion, and hardware-accelerated performance.请按 stitch-design-taste skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具sxtyj-frontend-dependency-repairstandards-stack/agent-skills/skills/sxtyj-frontend-dependency-repair/SKILL.mdLoad when working on sxtyj/public/static Vue frontend installs or builds, especially chromedriver timeouts, node-sass missing vendor binaries, or missing echarts world_new map modules.请按 sxtyj-frontend-dependency-repair skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉system-grounded-requirement-analysisstandards-stack/agent-skills/skills/system-grounded-requirement-analysis/SKILL.mdLoad when doing requirements analysis from Excel, screenshots, chat notes, or customer text and the user expects local-system screenshots, Before/After prototypes, solution depth, fast launch, or no-code-change judgment.请按 system-grounded-requirement-analysis skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具task-index-library-maintenancestandards-stack/agent-skills/skills/task-index-library-maintenance/SKILL.mdUse when tasks, Yunxiao work items, Codeup MRs, production log investigations, evidence packs, or user-taught workflows should be recorded in the project-wide unified task index.请按 task-index-library-maintenance skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具task-thread-title-ticket-prefixstandards-stack/agent-skills/skills/task-thread-title-ticket-prefix/SKILL.mdUse when creating, naming, renaming, or spawning a thread/subthread/subagent for a Yunxiao task, Codeup task, VWCG ticket, or other task with a stable ticket number.请按 task-thread-title-ticket-prefix skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉design-taste-frontendstandards-stack/agent-skills/skills/taste-skill/SKILL.mdAnti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-f...请按 design-taste-frontend skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
设计 / 视觉design-taste-frontend-v1standards-stack/agent-skills/skills/taste-skill-v1/SKILL.mdThe original v1 taste-skill, preserved for projects depending on its exact behavior. The current default is `design-taste-frontend` (v2 experimental), which is a substantial rewrite. Use this v1 install name only if you need exact backward compatibility.请按 design-taste-frontend-v1 skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agenttaste-skill-visual-directionstandards-stack/agent-skills/skills/taste-skill-visual-direction/SKILL.mdLoad when improving the visual taste of zhctprompt HTML, Word, PPT, product decks, manuals, reports, or UI prototypes, especially when the user says taste/test 品味, 美观, 整体改风格, 高级感, or asks Codex to choose the default visual style.请按 taste-skill-visual-direction skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentteam-document-review-learning-loopstandards-stack/agent-skills/skills/team-document-review-learning-loop/SKILL.mdBottom-layer workflow for code and document changes in zhctprompt: record intent and evidence as Markdown for AI, generate HTML for human review, capture human feedback, and convert approved practices into skills/knowledge while preserving rejected practice...请按 team-document-review-learning-loop skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentteam-weekly-report-generationstandards-stack/agent-skills/skills/team-weekly-report-generation/SKILL.mdUse when teammates need to turn scattered Chinese work notes into a unified weekly report for 康比特数字技术中心 / 数科业务, especially when the user says 写周报, 整理周报, 统一周报, 本周工作, 下周计划, or asks to generate reports from team work records.请按 team-weekly-report-generation skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全test-driven-developmentstandards-stack/agent-skills/skills/test-driven-development/SKILL.mdDrives development with tests. Use when implementing any logic, fixing any bug, or changing any behavior. Use when you need to prove that code works, when a bug report arrives, or when you're about to modify existing functionality.请按 test-driven-development skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具thinking-model-answeringstandards-stack/agent-skills/skills/thinking-model-answering/SKILL.mdLoad by default for non-trivial Q&A, analysis, recommendations, planning, reviews, strategy, product, management, communication, learning, or decision questions where an answer should first select suitable thinking models and then reason through them.请按 thinking-model-answering skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentuser-taught-skill-capturestandards-stack/agent-skills/skills/user-taught-skill-capture/SKILL.mdThis skill should be used when the user explicitly teaches a workflow, says "以后不需要我再提醒", "沉淀成 skill", "装进本项目", "复制给其他 agent 线程", "记住这个做法", or provides step-by-step instructions that should become reusable project behavior.请按 user-taught-skill-capture skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂wang-huiwen-demand-opportunitystandards-stack/agent-skills/skills/wang-huiwen-demand-opportunity/SKILL.mdLoad when evaluating whether a product demand, customer need, opportunity, timing window, or "why now" claim is real, broad, intense, feasible, ROI-positive, and grounded in infrastructure or behavior change.请按 wang-huiwen-demand-opportunity skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂wang-huiwen-founder-product-managerstandards-stack/agent-skills/skills/wang-huiwen-founder-product-manager/SKILL.mdLoad when mentoring, assessing, or redesigning product-manager work around entrepreneurial ownership, business-result responsibility, cross-domain learning, PM capability growth, or moving from feature delivery to product/business design.请按 wang-huiwen-founder-product-manager skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂wang-huiwen-market-strategystandards-stack/agent-skills/skills/wang-huiwen-market-strategy/SKILL.mdLoad when evaluating market size, scale effects, industry-chain concentration, high-frequency vs low-frequency behavior, incremental vs stock markets, entry timing, latecomer strategy, or competitive strategy for a product or business.请按 wang-huiwen-market-strategy skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品 / 市场 / 智慧食堂wang-huiwen-pmf-stp-4pstandards-stack/agent-skills/skills/wang-huiwen-pmf-stp-4p/SKILL.mdLoad when turning a product idea into PMF, launch wedge, innovation-diffusion stage, STP, 4P, pricing, channel, promotion, or go-to-market decisions.请按 wang-huiwen-pmf-stp-4p skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
验证 / 质量 / 安全webapp-testingstandards-stack/agent-skills/skills/webapp-testing/SKILL.mdToolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.请按 webapp-testing skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档wechat-article-youdao-intakestandards-stack/agent-skills/skills/wechat-article-youdao-intake/SKILL.mdLoad when the user sends a WeChat public-account article link and asks to save it to Youdao Note, archive it into zhctprompt, turn it into a project skill, update Codex memory, or make future agents remember the workflow.请按 wechat-article-youdao-intake skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具weekly-report-strategic-transparencystandards-stack/agent-skills/skills/weekly-report-strategic-transparency/SKILL.mdUse when a weekly report should raise strategic transparency instead of merely listing work, especially for executive-facing, cross-team, product strategy, delivery, or management weekly reports.请按 weekly-report-strategic-transparency skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
知识库 / 资料归档youdao-chat-wedrive-project-intakestandards-stack/agent-skills/skills/youdao-chat-wedrive-project-intake/SKILL.mdLoad when the user asks to fetch a Youdao or WeChat group-chat note and quickly create WeDrive project folders plus the software/hardware online tracking sheet.请按 youdao-chat-wedrive-project-intake skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-batch-subagent-executionstandards-stack/agent-skills/skills/yunxiao-batch-subagent-execution/SKILL.mdThis skill should be used when the user asks to process multiple unfinished Yunxiao tasks, filter work items by owner or assignee, "使用 subagent 处理每个任务", "负责人为某人 的任务", "批量处理云效任务", or gives a Yunxiao project/view URL that may contain more than one actionable...请按 yunxiao-batch-subagent-execution skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-cli-interactionstandards-stack/agent-skills/skills/yunxiao-cli-interaction/SKILL.mdUse this skill for any Yunxiao work item interaction, including reading, creating, updating, commenting, status transitions, time records, TestHub, milestones, versions, or batch intake. All Yunxiao actions must go through CLI or OpenAPI-backed CLI scripts...请按 yunxiao-cli-interaction skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-codeup-pr-associationstandards-stack/agent-skills/skills/yunxiao-codeup-pr-association/SKILL.mdUse when a Yunxiao work item, VWCG ticket, Codeup branch, commit, PR, MR, release pre-merge, or code association is involved; especially when the user says "代码 0", "没有关联任务号", "以 # 任务号开头并且有空格", "提交 PR", "没有提交的 PR", "MR 提到 release", "预上线分支", or "云效 Code review".请按 yunxiao-codeup-pr-association skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-requirement-task-lifecyclestandards-stack/agent-skills/skills/yunxiao-requirement-task-lifecycle/SKILL.mdUse when pulling or implementing a Yunxiao requirement, especially when the user says 拉取需求, 完成需求, 提交代码, 拆分任务, 填计划时间, 预计工时, 实际工时, 修改需求状态, or asks to handle a VWCG requirement end to end. It requires creating a child task, filling plan fields and estimated ef...请按 yunxiao-requirement-task-lifecycle skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-smart-canteen-requirementsstandards-stack/agent-skills/skills/yunxiao-smart-canteen-requirements/SKILL.mdUse when the user wants to add, organize, or confirm product requirements in Yunxiao for project 【自研】智慧营养健康餐厅, especially from screenshots or brief Chinese requirement notes. Creator is determined by the configured Yunxiao account/token; assignee must be co...请按 yunxiao-smart-canteen-requirements skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
云效 / Codeup / APIyunxiao-wedrive-agentic-collaborationstandards-stack/agent-skills/skills/yunxiao-wedrive-agentic-collaboration/SKILL.mdUse when the user asks to adapt AI multi-agent development collaboration into zhctprompt, especially code review agents, project-management/task-scheduling agents, Yunxiao kanban automation, or enterprise WeDrive-backed collaboration.请按 yunxiao-wedrive-agentic-collaboration skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具zhct-docker-php-verificationstandards-stack/agent-skills/skills/zhct-docker-php-verification/SKILL.mdUse when verifying PHP syntax, ThinkPHP commands, or local backend behavior for zhct store/ai_api on lqt's machine; the host shell may not have php, so run PHP checks directly in the local Docker container.请按 zhct-docker-php-verification skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具zhct-readme-docker-startstandards-stack/agent-skills/skills/zhct-readme-docker-start/SKILL.mdUse when the user asks to run the ZHCT Docker project, says to follow zhctprompt/README.md, asks why 8088 exists, or says the store docker-compose.yml should not be used.请按 zhct-readme-docker-start skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentzhct-root-agent-bootstrapstandards-stack/agent-skills/skills/zhct-root-agent-bootstrap/SKILL.mdUse when an agent starts from the zhct workspace root or when a project rule in zhctprompt/AGENTS.md failed to trigger because the control project was not loaded.请按 zhct-root-agent-bootstrap skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
通用工程 / 本地工具zhctproject-repo-syncstandards-stack/agent-skills/skills/zhctproject-repo-sync/SKILL.mdLoad when the user says "拉取最新代码", "更新代码", "同步代码", or asks to prepare a zhct workspace on another computer; sync every repository registered under zhctproject, cloning missing repos and pulling existing repos.请按 zhctproject-repo-sync skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentzhctprompt-online-file-linkstandards-stack/agent-skills/skills/zhctprompt-online-file-link/SKILL.mdLoad when the user asks for a zhctprompt file's online access link, deployed URL, file link, MD/HTML viewing URL, or says 查询文件连接/文件链接 for files in this AI collaboration project.请按 zhctprompt-online-file-link skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
团队治理 / Agentzhctprompt-self-update-on-qastandards-stack/agent-skills/skills/zhctprompt-self-update-on-qa/SKILL.mdLoad before substantive Q&A or task execution in the zhct workspace to update the zhctprompt control project from Codeup with git pull --ff-only, while stopping on local changes, non-Git paths, authentication failures, or non-fast-forward state.请按 zhctprompt-self-update-on-qa skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-ceo-attention-standardstandards-stack/agent-skills/skills/creative-selection-ceo-attention-standard/SKILL.md当用户在领导团队、创始人管理、CEO 角色、追求卓越、是否该深入细节或如何设定组织标准时使用;强调 CEO 关注什么,公司就重视什么,所有职能都可能影响客户体验。不适用于把领导者变成事必躬亲的长期微管理。请按 creative-selection-ceo-attention-standard skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-crisis-responsestandards-stack/agent-skills/skills/creative-selection-crisis-response/SKILL.md当用户或团队遇到产品事故、客户风险、舆情、质量、安全、组织突发灾难或成长阵痛时使用;先解决问题、领导深入细节、听取建议、持续沟通、对客户影响负责并道歉。不适用于日常小故障或尚未确认的传闻扩散。请按 creative-selection-crisis-response skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-customer-journey-prototypestandards-stack/agent-skills/skills/creative-selection-customer-journey-prototype/SKILL.md当用户在做产品、服务、交付包、营销或客户支持设计时使用;把从发现、理解、购买、安装、使用、维护、退换到忠诚的完整客户旅程做成可见原型。不适用于只需修复一个独立 UI 缺陷且整体旅程已清楚的微小改动。请按 creative-selection-customer-journey-prototype skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-data-opinion-decisionstandards-stack/agent-skills/skills/creative-selection-data-opinion-decision/SKILL.md当用户纠结一个决策到底该继续收集数据、做测试,还是需要基于洞察和愿景拍板时使用;适用于创新产品、路线图、设计取舍、老板要更多数据但数据无法证明未来的场景。不适用于已有充分事实、可直接计算优化的纯数据问题。请按 creative-selection-data-opinion-decision skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-design-everythingstandards-stack/agent-skills/skills/creative-selection-design-everything/SKILL.md当用户把设计误认为只是视觉、UI 或设计师职责,或者想改善流程、命名、包装、退货、交付、内部工具等体验时使用;通过客户、痛点、约束、原型和去习惯化来重新设计问题。不适用于只要求专业视觉出稿且需求已清晰的纯美术任务。请按 creative-selection-design-everything skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-disrupt-evolve-executestandards-stack/agent-skills/skills/creative-selection-disrupt-evolve-execute/SKILL.md当用户要规划 V1/V2/V3、判断何时颠覆、何时进化、何时自我颠覆,或担心创新过度/不足时使用;要求 V1 有清晰颠覆点,V2 后用真实数据进化,并持续寻找自我颠覆窗口。不适用于只有维护性优化的常规版本计划。请按 creative-selection-disrupt-evolve-execute skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-failure-learning-loopstandards-stack/agent-skills/skills/creative-selection-failure-learning-loop/SKILL.md当用户在选择工作、创业方向、早期职业路径或高不确定项目时,强调通过实践、失败、复盘来学习;适用于问“我该不该冒险”“失败后怎么吸收教训”“怎么选能让我成长的机会”。不适用于单纯安慰、励志口号或已需要严密风控的高代价决策。请按 creative-selection-failure-learning-loop skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-growth-breakpoint-designstandards-stack/agent-skills/skills/creative-selection-growth-breakpoint-design/SKILL.md当用户的团队从 15 到 50、120、350 人等阶段增长,出现沟通失灵、职责混乱、会议过载、文化稀释或 IC 转管理者困难时使用;要求提前 2-3 个月设计组织、沟通、HR、教练和文化机制。不适用于小团队短期冲刺中的普通协作摩擦。请按 creative-selection-growth-breakpoint-design skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-heartbeat-constraintsstandards-stack/agent-skills/skills/creative-selection-heartbeat-constraints/SKILL.md当用户要让不确定项目真正推进、设定 V1 截止日期、安排内部里程碑、避免无休止打磨或版本节奏混乱时使用;强调外部心跳、内部心跳、团队心跳和项目心跳。不适用于探索初期仍未定义问题空间的纯研究。请按 creative-selection-heartbeat-constraints skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-hiring-three-crownsstandards-stack/agent-skills/skills/creative-selection-hiring-three-crowns/SKILL.md当用户要设计招聘流程、面试候选人、扩张团队、避免文化稀释或判断候选人是否适合跨职能协作时使用;强调招聘经理加两个内部客户管理者共同决策、工作模拟、文化接种和不招有毒人才。不适用于临时外包或低协作岗位的快速采购。请按 creative-selection-hiring-three-crowns skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-obsessive-idea-filterstandards-stack/agent-skills/skills/creative-selection-obsessive-idea-filter/SKILL.md当用户有创业/产品/业务想法,想判断它是杂念、维生素还是止痛药,是否值得投入数月到数年时使用;要求先延迟直觉、研究原型、识别致命风险,再看想法是否仍挥之不去。不适用于低成本日常优化或已验证市场的版本迭代。请按 creative-selection-obsessive-idea-filter skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-pm-customer-voicestandards-stack/agent-skills/skills/creative-selection-pm-customer-voice/SKILL.md当用户要定义产品经理职责、拆分 PdM/PjM/PgM、建立产品管理流程,或需要有人在工程、设计、营销、销售、支持和法律之间维护客户声音时使用。不适用于只需要项目排期、会议纪要或任务跟踪的项目管理场景。请按 creative-selection-pm-customer-voice skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-product-story-whystandards-stack/agent-skills/skills/creative-selection-product-story-why/SKILL.md当用户需要讲清产品、项目、公司或方案为什么存在,如何说服客户、团队、投资人、领导或合作伙伴时使用;强调先讲 why,再讲 what,并把故事作为产品定义工具。不适用于包装一个尚未解决真实问题的空洞卖点。请按 creative-selection-product-story-why skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-revolutionary-opportunity-filterstandards-stack/agent-skills/skills/creative-selection-revolutionary-opportunity-filter/SKILL.md当用户要判断一家公司、项目、赛道或工作机会是否值得投入青春和精力时使用;看它是否组合新技术、解决真实痛点、时机成熟、领导愿意转向、并以新方式理解用户。不适用于只比较薪资、福利或品牌名气的普通求职筛选。请按 creative-selection-revolutionary-opportunity-filter skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-single-customer-rulestandards-stack/agent-skills/skills/creative-selection-single-customer-rule/SKILL.md当用户在 B2B/B2C/B2B2C、平台、多边市场或产品定位中试图同时讨好多类客户时使用;要求选定一个真正主子,让产品、品牌、销售、支持、定价围绕它组织。不适用于多个内部利益相关者的普通项目管理。请按 creative-selection-single-customer-rule skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。
产品方法论 / 创造creative-selection-up-and-around-scanstandards-stack/agent-skills/skills/creative-selection-up-and-around-scan/SKILL.md当用户或团队陷入局部任务、只盯当前交付、担心项目方向走偏或想从 IC 成长为管理者时使用;要求定期向前看使命和里程碑,向周围看其他职能和内部客户。不适用于需要短时间专注冲刺且方向已被充分校准的局部任务。请按 creative-selection-up-and-around-scan skill 处理这件事;我会提供背景、资料路径、期望输出和验收标准。

16. Skill 到底有几种用法

新手不要先背所有 skill 名称。先理解:skill 是让 Agent 不从零摸索的可复用工作方法。它既可以是项目规则,也可以是本机能力、任务路由、方法论、交付流程或团队治理规则。

方式 放在哪里 什么时候用 新手可以怎么说 典型例子
项目内置 skillstandards-stack/agent-skills/skills/zhctprompt 里执行项目任务,必须按团队规则来。按本项目规则处理这个任务。zhctprompt-self-update-on-qaproduct-strategy-planning
本机 Codex skill$HOME/.codex/skills/让同事在自己的 Codex 线程里也能调用项目沉淀的方法。把项目 skills 安装到本机 Codex。codex-project-skill-auto-install 和安装脚本
任务路由 skill项目内 skill 或本机 skill帮 Agent 判断这件事该走产品、开发、云效、微盘、手册还是视频流程。先判断这个任务适合哪个流程。product-development-master-agentsmart-canteen-agent-native-work-system
方法论 / 知识库 skillstandards-stack/llm-wiki/ + 对应 skill把书、课程、文章、会议材料转成可复用判断方法。把这套方法沉淀成以后能复用的技能。王慧文产品课方法组、llm-wiki-knowledge-base
交付 / 证据 skill项目内 skill + work*/summary.md做完功能、手册、视频、截图、验收材料,要求证据闭环。做完以后把手册、截图、验证证据都补齐。post-test-manual-video-deliveryoperation-manual-video
治理 / 团队协作 skill项目内 skill + AGENTS.md / tasks/QUEUE.md维护规则、任务索引、团队交接、review 学习闭环。这个经验以后团队都按这个走。codex-team-skill-handoffteam-document-review-learning-loop
外部导入 / 可选 skill先评审,再安装到本机或项目现有项目 skill 不够用,需要引入第三方或个人 skill。先看看有没有合适 skill,安装前审一下风险。find-skillsskill-vetter、设计类可选 skill
最简单的判断:

一次性资料说明写 Markdown / HTML;会重复发生、容易出错、需要 Agent 以后自动遵守,才考虑做成 skill。团队都要遵守的放进项目 skill,只有个人常用的放进本机 Codex skill。

17. 五模块怎么和日常工作对应

五模块谁主要用日常例子常见输出
场景产品化产品、交付、项目经理把客户反馈整理成角色、流程、痛点、验收指标场景说明、业务流程、验收指标
软件平台化产品、研发、测试PC、小程序、APP、API、权限、菜单、数据流页面说明、接口契约、测试用例
硬件体系化设备、研发、交付点餐机、取餐柜、一体机、摄像头、秤、屏设备清单、协议、联调记录
AI 工程化Agent owner、研发、产品Prompt、MCP、代码索引、AI review、自动化测试skill、评估表、知识库、工具链
交付资产化交付、产品、测试、销售上线手册、培训、演示、复盘、客户说明手册、Word、截图、视频、公告

18. 给新人直接复制的提问模板

我要加入这个项目协作,请先帮我判断我的任务归属。

我要做的事:
资料来源:
属于产品类还是项目类:
涉及客户/项目/产品:
目标用户:
期望输出:
验收标准:
不能做的事:

请你先告诉我:
1. 应该读哪些入口文件;
2. 应该放到哪个目录;
3. 应该使用哪个微盘索引或项目库;
4. 应该使用哪些项目 skill;
5. 最后需要留下哪些证据。

19. 不能做什么

20. 最终记忆点

第一,zhctprompt 是团队 AI 协作控制项目,不是业务代码仓库。
第二,资料先分产品类和项目类:产品进 modules/product,项目进 modules/project 和微盘索引。
第三,任何有价值的工作都要留下可恢复证据:Markdown 给 AI 继续,HTML 给人 review,索引给团队复用。

来源口径:README、PROJECT_IPO_USAGE_GUIDE、modules、work_company_knowledge、AI Agent 入门介绍、会议与 AI 协作流程规范、项目 skill registry,以及《创造》产品创造方法组。全量清单来源:work/2026-06-09-beginner-skill-usage-guide/project-skill-usage-inventory.csv

21. 版本记录

版本更新日期更新内容数据快照
V1.12026-07-23新增八类可复用沉淀、四个代表性落地,刷新资产数量,并把 187 项 skill 长表明确标为历史说明快照。2026-07-23
V1.02026-07-23新增阿里云运维 Agent 员工协作流程、角色边界、请求模板和生产变更门禁。阿里云验收证据

后续更新规则:内容或数据口径变化升级次版本;仅修正文案或链接升级补丁版本;页面结构或使用对象发生重大变化时升级主版本。每次必须同时写明版本、更新日期、数据快照和本次更新摘要。