Token导航 LogoToken导航TokenDH.com
开发只读github未标认证来源可访问许可证需确认审计通过

diataxisdiataxis 搜索

Agent Skill

diataxis 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

423

周安装

18

GitHub Stars

5

下载量

148
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:diataxis(diataxis 搜索)
来源仓库:https://github.com/keithpatton/diataxis-agent-skill
仓库路径:skills/diataxis
安装命令:
npx skills add https://github.com/keithpatton/diataxis-agent-skill --skill diataxis
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/keithpatton/diataxis-agent-skill --skill diataxis

简介

diataxis 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态进行整理时使用。

  • 适用于文档框架分类、内容审计与知识体系构建类开发支持场景。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,需确认权限与网络访问能力。
  • 使用前建议核实维护状态及是否涉及文件读写或外部依赖调用。
  • diataxis 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Diátaxis Documentation Governance

This skill enforces the Diátaxis framework for documentation. Diátaxis identifies four distinct documentation types based on user needs: tutorials, how-to guides, reference, and explanation. Each serves a different purpose and must be written differently.

Operating Modes

This skill operates in one of the following modes:

  • classify: Identify which quadrant content belongs to
  • audit: Analyze a doc set for gaps, imbalances, and violations
  • generate: Create new documentation in a specified quadrant
  • restructure: Split or reorganize existing mixed-mode content

The mode must be inferred from the user request or explicitly stated. If the operating mode cannot be confidently inferred, ask for clarification before proceeding.

Refusal with explanation is a valid and successful outcome when quadrant purity cannot be maintained.

Generate Mode Policy

For generate mode, one of the following must be true before producing content:

  • The user explicitly states the target quadrant ("write a How-to for X"), OR
  • The agent classifies the request, presents reasoning with confidence+evidence, and receives user confirmation

The Diátaxis Compass

Use this decision tree to classify content into exactly ONE quadrant:

If content......and serves user's......then it belongs to...
informs actionacquisition (study)Tutorial
informs actionapplication (work)How-to Guide
informs cognitionapplication (work)Reference
informs cognitionacquisition (study)Explanation

Two questions to ask:

  1. Does this inform action (doing) or cognition (knowing)?
  2. Does this serve acquisition (learning/study) or application (working)?

Quadrant Summary

Tutorial (learning-oriented)

A lesson that takes the learner through a practical experience. The instructor is responsible for the learner's success. Focus on doing, not explaining. Deliver results early and often. Minimize explanation. No choices or alternatives.

  • Form: Sequential lesson with concrete steps
  • Language: "We will...", "First, do X", "You'll notice that..."
  • Not: Teaching concepts, offering alternatives, explaining why

How-to Guide (goal-oriented)

Directions that guide an already-competent user through a real-world problem. Assumes the reader knows what they want to achieve. Focused on work, not study.

  • Form: Series of steps addressing a specific goal
  • Language: "To achieve X, do Y", "If you want...", conditional imperatives
  • Not: Teaching beginners, explaining concepts, describing machinery

Reference (information-oriented)

Technical description of the machinery. Austere, neutral, accurate. Structure mirrors the thing being described. The user consults it, not reads it.

  • Form: Structured descriptions, tables, specifications
  • Language: "X is...", "The options are...", factual statements
  • Not: Instructing, explaining why, narrative, opinions

Explanation (understanding-oriented)

Discursive treatment that provides context, background, and answers "why". Makes connections, admits opinion, circles around the subject.

  • Form: Prose discussion of a topic
  • Language: "The reason for X is...", "This is because...", "Consider..."
  • Not: Step-by-step procedures, technical specifications, teaching tasks

Core Workflow

For Classification

  1. Read the content carefully
  2. Apply the compass: action/cognition × acquisition/application
  3. Identify the single quadrant
  4. Report with confidence and evidence

For Generation

  1. Confirm the target quadrant (stated or inferred with confirmation)
  2. Apply quadrant constraints strictly
  3. Refuse to blend quadrants; recommend splitting if needed
  4. Validate output against quadrant characteristics

For Audit

  1. Inventory all documentation
  2. Classify each document
  3. Identify gaps (missing quadrants)
  4. Flag violations (mixed-mode, wrong quadrant)
  5. Report imbalances (e.g., "reference-heavy, tutorial-poor")

For Restructure

  1. Classify the existing content
  2. Identify quadrant violations
  3. Propose splits (one document per quadrant)
  4. Present plan for user confirmation before changes

Output Requirements

When classifying or flagging violations, always provide:

  • Quadrant: The identified type (Tutorial, How-to, Reference, Explanation)
  • Confidence: high | medium | low
  • Evidence: Specific phrases, structural patterns, or signals

Example:

Classified as How-to (high confidence). Evidence: imperative verbs ("configure", "set up"), goal-oriented heading, absence of conceptual framing, assumes reader competence.

Violation Detection

Flag these anti-patterns:

  • Tutorial with explanation: Lengthy "why" sections, conceptual digressions
  • How-to that teaches: Beginner framing, "let's learn" language
  • Reference with narrative: First-person voice, procedural instructions
  • Explanation with procedures: Numbered steps, imperative commands
  • Mixed-mode documents: Multiple quadrant signals in single document

For detailed detection patterns and remediation, see references/anti-patterns.md.

Non-Goals

This skill does not:

  • Invent documentation strategy or information architecture
  • Decide product architecture or feature scope
  • Merge multiple Diátaxis quadrants into a single document
  • Override repository-specific documentation rules or style guides
  • Generate content without explicit quadrant classification

Additional Resources

Attribution

Diátaxis is the work of Daniele Procida. This skill encodes the Diátaxis framework for use by AI agents. For the authoritative source and complete documentation, see diataxis.fr.

Licensed under CC-BY-SA 4.0. Citation metadata available at the Diátaxis GitHub repository.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35.92%
按下载量换算53

Claude

28.99%
按下载量换算43

Cursor

17%
按下载量换算25

Gemini CLI

8.53%
按下载量换算13

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills