一键导入
diataxis-docs
Use when updating, rewriting, or auditing documentation to follow the Diataxis framework with its four distinct content types
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when updating, rewriting, or auditing documentation to follow the Diataxis framework with its four distinct content types
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | diataxis-docs |
| description | Use when updating, rewriting, or auditing documentation to follow the Diataxis framework with its four distinct content types |
Restructure documentation using the Diataxis framework - a systematic approach that organizes content into four distinct types based on user needs: Tutorials, How-to guides, Reference, and Explanation.
PRACTICAL THEORETICAL
┌─────────────────────┬─────────────────────┐
│ │ │
LEARNING │ TUTORIALS │ EXPLANATION │
(study) │ learning-oriented │ understanding-oriented
│ │ │
├─────────────────────┼─────────────────────┤
│ │ │
WORK │ HOW-TO GUIDES │ REFERENCE │
(apply) │ goal-oriented │ information-oriented
│ │ │
└─────────────────────┴─────────────────────┘
Purpose: Take the user through a learning experience User question: "Can you teach me to...?"
| Do | Don't |
|---|---|
| Let the user learn by doing | Explain concepts in depth |
| Get them started immediately | Offer choices or alternatives |
| Ensure every step works | Focus on edge cases |
| Show concrete results early | Assume prior knowledge |
Language: "In this tutorial, we will..." / "First, do x. Now, do y."
Purpose: Help the user accomplish a specific task User question: "How do I...?"
| Do | Don't |
|---|---|
| Focus on the specific goal | Teach or explain |
| Provide numbered steps | Cover every option |
| Address real-world problems | Start from first principles |
| Be adaptable to variations | Include unrelated information |
Language: "This guide shows you how to..." / "If you want x, do y."
Purpose: Describe the machinery and how to operate it User question: "What is...?" / "What does X do?"
| Do | Don't |
|---|---|
| Be accurate and complete | Explain why or how to use |
| Follow consistent structure | Include tutorials or guides |
| Mirror the code/system structure | Offer opinions |
| Provide examples of usage | Describe user journeys |
Language: "X is configured by..." / "The options are: a, b, c."
Purpose: Clarify and illuminate a topic User question: "Why...?" / "Can you explain...?"
| Do | Don't |
|---|---|
| Provide context and background | Provide step-by-step instructions |
| Discuss alternatives | Describe the machinery |
| Explain design decisions | Focus on a single correct way |
| Connect to broader concepts | Be exhaustive about details |
Language: "The reason for x is..." / "An alternative approach is..."
For each document, ask:
Create an inventory:
| Document | Current Type | Issues | Target Type |
|----------|--------------|--------|-------------|
| README.md | Mixed | Tutorial + Reference mixed | Split |
| setup.md | How-to | Contains explanation | Extract |
Mixed content signals:
Split rule: Each document serves ONE quadrant.
docs/
├── tutorials/ # Learning experiences
│ ├── getting-started.md
│ └── first-project.md
├── how-to/ # Task completion
│ ├── configure-x.md
│ └── deploy-to-y.md
├── reference/ # Technical descriptions
│ ├── api.md
│ ├── config-options.md
│ └── cli-commands.md
└── explanation/ # Understanding
├── architecture.md
└── design-decisions.md
For each document:
Cross-reference pattern:
> For a tutorial on getting started, see [Getting Started](../tutorials/getting-started.md).
> For details on all configuration options, see [Configuration Reference](../reference/config.md).
| Type | Orientation | Serves | Structure |
|---|---|---|---|
| Tutorial | Learning | Study | Sequential lessons |
| How-to | Goal | Work | Numbered steps |
| Reference | Information | Work | Consistent descriptions |
| Explanation | Understanding | Study | Discursive prose |
| Mistake | Fix |
|---|---|
| How-to that teaches | Extract learning content to tutorial |
| Tutorial with all options | Remove options, keep one path |
| Reference with "why" | Move explanations to explanation docs |
| Explanation with steps | Extract steps to how-to guide |
| README that does everything | Split into appropriate documents |
Research and recommend the best local LLM model stack for the current machine. Detects hardware (CPU, GPU, RAM, VRAM), discovers installed hosting stacks (Ollama, llama.cpp, vLLM, etc.), searches Hugging Face for compatible models, and enables easy downloads. Use when the user wants to find, evaluate, or download local AI models that will run well on their system.
Zellij terminal multiplexer — creating KDL layouts, managing sessions via CLI, and running commands without disrupting the user's workspace
Use when adding a new service to a Nix flake — creating packages from upstream repos (Go, Node, Python), writing service modules, managing configs, wiring secrets, and testing. Covers the full lifecycle from source to running daemon.
Use when a launchd service on nix-darwin fails to start, exits with non-zero status, or doesn't restart after a switch. Triggers on: exit code 78, "EX_CONFIG", "Operation not permitted", daemon vs user.agent decisions, plist not being reloaded.
Use when managing 1Password secrets via Nix on nix-darwin. Triggers on: adding API keys, tokens, or credentials to service configs; wiring opnix 1Password items into home-manager modules; creating opnix-managed secret files with activation script ordering.
Use when tracking, planning, implementing, or reviewing work using yx (yaks) with the autonomous /shave loop, or when multiple agents need to coordinate on shared tasks