User Manual Authoring logo

User Manual Authoring

Organization
devcodex-labs
user-manual-authoring

开源/公开用户站点与最终用户手册写作 Owner — guide/README/reference/migration/changelog/operations;与维护者开发站不等价;经 DocsAudienceIntent 路由。

Overview

Publisherdevcodex-labs
Repositorydevcodex
Skill nameuser-manual-authoring
Stars
433
Forks
44
Bundled files
1
LicenseAGPL-3.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by devcodex-labs on GitHub. Read the source before you install it.

Installation

Install the User Manual Authoring AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/devcodex-labs/devcodex/tree/main/content/skills/user-manual-authoring
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/devcodex-labs/devcodex.git /tmp/devcodex
mkdir -p .claude/skills
cp -r /tmp/devcodex/content/skills/user-manual-authoring .claude/skills/user-manual-authoring
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable User Manual Authoring in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads User Manual Authoring on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so User Manual Authoring is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

User Manual Authoring Skill

职责

当任务目标是开源/公开用户使用站点文档、最终用户手册、README、quick start、接入手册、API/CLI reference(用户侧契约面) 或公开能力页时,本 Skill 是优先写作入口。

主受众必须是使用者(含「使用库的开发者」),不是本仓库维护者。成功标准:理解 → 安装/接入 → 第一次成功 → 任务 → 配置 → 失败恢复。

技术方案、接口契约深页、实现计划和维护者 checklist 只能后置或拆任务;维护者开发站走 maintainer-docs-site-authoring。

触发条件

场景是否触发
docsAudience=public-user(DocsAudienceIntentGate)必须
用户要求“用户使用文档 / 使用手册 / 最终用户文档 / 开源用户站”必须
用户要求 README、quick start、接入手册、官网使用说明或公开能力页必须
docs-first、先写文档再开发、后续按文档实现必须
API/CLI/Config reference(使用者查表)必须(可编排 dev-docs light-api)
维护者开发站 / contributing / 发版 runbook / ADR 主叙事N/A → maintainer-docs-site-authoring
仅「写文档站/website」且 ambiguousN/A → 阻断消歧,禁止开写
纯架构说明且受众为维护者N/A → dev-docs / maintainer-docs-site-authoring

输入顺序

用户最终文档不得直接从未确认草稿生成最终承诺。推荐顺序:

  1. 原始需求或材料保留为输入锚点。
  2. 00-需求概况.md 只作为整理草稿,不替代确认事实源。
  3. 以 01-需求确认.md 或产品直接提供的 01-产品需求.md 作为文档事实源。
  4. 判断 documentationSurface=docs-site | README-minimum | N/A。
  5. 先产出最终用户使用文档。
  6. 涉及前端、API 或外部调用方时,再生成契约文档。
  7. 技术方案、实施方案、实施计划和 ECR 必须对照用户文档。

文档契约

字段要求
primaryAudience必须是用户 / 使用者;内部开发者只能是次级受众
docsSurfaceguide | readme | reference | migration | changelog | operations(由 DocsAudienceIntent 锁定)
documentLocation写清是文档站、README、需求交付目录、官网页还是 maintainer-only
userJourney覆盖理解、安装/进入、第一次成功、常见任务、配置、失败处理和下一步
informationArchitecture文档站要区分用户手册、reference、operations、compatibility、implementation、maintainer
pageRoleMatrix多页文档站列出页面 role、受众、sourceOfTruth、nav/sidebar 位置和用户主路径状态
sidebarSemanticModel文档站列出每个 sidebar group 的用户任务模型、相邻页面职责、route/label 真相源和非归属说明
configurationModel配置字段、默认值、选择建议、错误与排错必须简单易懂
realWorkflowExample队列、任务、异步、导入导出、推送或批处理类文档必须给真实批量工作流
renderedFlowEvidenceMermaid / 流程图 / 文档站主题必须有真实渲染或运行态验证证据
semanticParityEvidence行为语义、默认值、兼容路径、支持/不支持承诺必须有 BehaviorSemanticDocsParityGate / NegativeTranslationParityProbe 证据
exampleTruthEvidenceoption/config/method/callback 示例必须有 DocsExampleTruthSurfaceGate / CallbackExampleScopeProbe 证据
expertOutputQualityEvidence示例、fixture、quick start 或接入手册必须有 expert-output-quality / ExpertOutputQualityGate 证据,区分生产推荐路径、框架原生能力、fixture/mock/demo 边界和反模式
developerInfoPlacement开发契约、技术方案、数据模型、维护者 checklist 后置或单独标记
consumerMap同步 README、website、Profile、examples、prompts、templates、validate 和部署副本

必执行门禁

共享 Gate 分组与 requiredEvidence 以 ../spec-governance/gate-registry.json 为准(user-manual / docs-ia-readability / docs-semantics-examples / docs-audience-render-sequence / scenario-durable-workflow / expert-output-quality)。下表是本 Owner 的差分清单(触发 → 要证什么),不维护跨 Skill 百科。

触发要证什么(摘要)
任意用户手册 / README / 站点文档冻结用户主受众与落点;主路径覆盖理解→第一次成功→任务→配置→失败恢复;公开面不混维护者 checklist
docs-first / 最终用户手册文档先于技术方案成为用户路径合同;写目标版本可执行路径,不是 preview 状态说明
文档站 / 多页 READMEpage role / sidebar 任务模型 / IA 分区;中文主表达;主题与生成站点运行态(不仅 Markdown)
队列 / 异步 / 批处理 quick start真实业务工作流;声称场景完整或持久化编排时追加 scenario / durable 证据(见 registry)
行为承诺 / 示例 / 回调语义与 public API/runtime 对齐;示例可追溯 type/schema/dispatcher;fixture 不得冒充生产推荐路径(expert-output-quality)
能力 / 导航 / 路径变更同步 README、website、Profile、examples、validate、部署副本与代码消费点

执行步骤

  1. 判定文档目标:docs-site、README-minimum、public page、requirement deliverable 或 N/A。
  2. 核对事实源:确认需求/产品需求、当前版本、发布状态和公开能力边界。
  3. 写出用户主路径:这是什么、适合谁、第一次成功、常见任务、配置、排错、限制、下一步。
  4. 拆分信息架构:用户手册与 API/CLI/config reference、operations、implementation、maintainer 分开。
  5. 建立 pageRoleMatrix 与 sidebarSemanticModel:确认每页 role、sidebar group、相邻页面职责、route 真相源和非归属说明。
  6. 对真实工作流补示例:避免只有单点 API 或单个硬编码 job。
  7. 对示例 / fixture / quick start 执行 expert-output-quality:冻结 roleBaseline、productionRecommendedPath、frameworkNativeCapability、fixtureBoundary、antiPatternContrast 与 evidenceMatrix。
  8. 建立 consumerMap,列出 README、website、Profile、examples、prompts、templates、validate、部署副本和代码消费点。
  9. 完成后按风险调用 audit-user-manual 做用户侧文档聚合 review;落点为 README / 主入口文档时再叠加 audit-readme,通用结构与准确性由 audit-document 承接。

与其他 Skill 的关系

  • dev-docs:识别文档任务后,用户站/README/用户手册必须 handoff 本 Skill;reference 可编排 light-api;纯维护者技术文或架构走 dev-docs / maintainer-docs-site-authoring。
  • maintainer-docs-site-authoring:维护者开发站 Owner;受众正交。
  • readme-authoring:README 是本 Skill 的 README 专项分支,继续负责 README 章节顺序与用户旅程细化。
  • audit-user-manual:负责用户侧文档、项目文档、菜单导航和文档 IA 的聚合审查,不替代写作入口。
  • audit-readme / audit-document:负责 README 专项与通用文档维度,通常由 audit-user-manual 编排。
  • expert-output-quality:负责专家型产物质量,避免文档把 fixture/mock/demo 或低阶重复写法包装成生产推荐路径。
  • test-router:选择生成站点、链接、用户路径、Browser/截图或代码级替代验证。
  • document-sync:按 consumerMap(含 audience=public-user)检查当前消费者和部署副本。
  • DocsAudienceIntentGate:模型判定受众,scripts/lib/docs-audience-intent.js 校验结构化决策;npm run test:docs-audience 仅验证契约接线。

认知高度与任务语言(L3 · 强制)

受众对(public-user)不等于可读。guide/readme 不得做成「完整但全是底层函数」的符号说明书。

认知高度三层

高度含义guide/readme 主路径reference
Task用户要完成的事、场景、步骤、选择建议必须为主叙事可附「何时用」
Concept领域概念、配置含义(白话)服务 Task可简要
Symbol类型名、方法签名、内部模块后置或链到 reference允许密

写作硬规则

  1. 唯一推荐路径:quick start 只推广一条 productionRecommendedPath;底层装配标「高级/扩展」。
  2. 先任务后符号:目录与标题优先任务名(「发消息」),不是 MessageDispatcher。
  3. 用户任务为主线:模型审查快速开始能否帮助目标读者完成第一次实际使用;不能用函数调用数量或是否出现某个任务词代替可读性判断。
  4. 渐进披露:5 分钟会用 → 30 分钟会选 → 查表 reference;禁止一篇写穿全部 public 函数当使用文档。
  5. 术语:内部名首次出现必须白话;配置先默认与选择建议再字段表。
  6. reference:符号可密,每项至少「用途一句话 + 与推荐路径关系/何时不用」。

延展失败场景(写作与审查时主动对照)

用户只提一种「看不懂」时,仍应自检相邻风险(问题驱动场景延展):

组场景
A 叙事函数清单 guide、概念堆无任务、配置字典、错误码无恢复、多入口无推荐
B IAguide/reference 混主路径、源码侧栏、深链才到第一次成功、一篇写穿
C 示例不可跑、fixture 当生产、无失败路径、版本漂移
D 伪用户库文档按贡献者写、运维当研发、术语三套
E 假完整TBD、超版承诺、三口径
F 负担前置未声明、图文不符
G 元失败受众对高度错、好读但假、单页好整站乱

完成前漂移自检

  • 锁定 docsAudience=public-user 后,正文不得以 release checklist / monorepo 内部 / ADR 列表 / 内部台账为首屏主叙事。
  • 模型审查读者任务、叙事顺序与可读性,形成绑定正文摘要的 DocsContentReviewV1(kind、conclusion、rationale、evidenceRefs,受众漂移追加 audience);适配函数只投影该结论,没有审查时为 unverified。
  • classifyDocsAudienceDriftSample('public-user', body, review) 与 classifyUserDocsCognitiveAltitudeSample(body, { surface, review }) 必须使用相应 kind 的当前审查记录;修改正文后重新审查。
  • 无安装/第一次成功路径不得宣称用户站完成。
  • 验证:模型内容审查、实际用户路径;npm run test:docs-audience 仅证明协议正确,不能证明某份文档可用。

禁止

  • 禁止把用户要求的站点文档写成开发文档、技术方案、实现计划或维护报告。
  • 禁止让开发契约、目标 API、Redis/缓存模型、数据模型或验收 checklist 成为用户主入口。
  • 禁止把未确认的 00-需求概况.md 当成最终用户承诺。
  • 禁止把最终用户手册写成整站全部内容容器。
  • 禁止把当前不可用状态说明冒充目标版本最终用户手册。
  • 禁止把 fixture、mock、demo、硬编码单例或重复 route/middleware/resource 声明写成用户主路径的生产推荐实践;必须标明验证用途和推荐替代。
  • 禁止在单任务内同时写维护者站并宣称双受众完成;多受众须拆任务。
  • 禁止以底层函数/类型调用链作为 quick start 的唯一主叙事(无任务句、无推荐路径)。

同步锚点(validate / consumer)

FixtureBoundaryDisclosureGate · ScenarioCoverageMatrixProbe · DurableBatchOrchestrationProbe

额外同步锚点

ChinesePrimaryExpressionGate · SidebarPageRoleMaterializationProbe · SidebarGroupSemanticModelProbe

UserFacingDeliveryChainGate · FinalUserManualFirstGate · UserPathContractSweep · UserManualProductizationGate · UserManualRenderedFlowAndRealWorkflowProbe · DocsPageRoleMatrixGate · DocsThemeRuntimeVisualProbeGate

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the User Manual Authoring AI skill do?

开源/公开用户站点与最终用户手册写作 Owner — guide/README/reference/migration/changelog/operations;与维护者开发站不等价;经 DocsAudienceIntent 路由。

Why use User Manual Authoring on TypingMind?

Because you install it once and use it with any model. User Manual Authoring is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install User Manual Authoring in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/devcodex-labs/devcodex/tree/main/content/skills/user-manual-authoring. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use User Manual Authoring?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with User Manual Authoring?

As many as you like. As long as a model supports skills, you can use User Manual Authoring with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the User Manual Authoring AI skill free?

Yes. It is published on GitHub by devcodex-labs under the AGPL-3.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇