Token导航 LogoToken导航

Agent Skill · 研究检索

documentation(文档)

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

documentation 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 72,323。

总安装 188周安装 8GitHub Stars 142下载 66核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

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:documentation(文档)
来源仓库:https://github.com/thebushidocollective/han
仓库路径:skills/documentation
安装命令:
npx skills add https://github.com/thebushidocollective/han --skill documentation
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/thebushidocollective/han --skill documentation

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论。涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

通过 npx skills add 命令从指定 GitHub 路径安装并使用。

来源仓库内路径:skills/documentation

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/thebushidocollective/han --skill documentation;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

documentation 怎么安装?

在终端执行 npx skills add https://github.com/thebushidocollective/han --skill documentation。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

documentation 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

documentation 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

documentation 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

documentation 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 142 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

documentation 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
documentation

文档 · 本页

188142接触敏感数据MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Documentation Skill

Create clear, useful documentation that helps people understand and use your code.

Core Principle

Write for the reader, not for yourself.

Types of Documentation

1. README Files

Purpose: Help users understand what the project does and how to use it

Audience: New users, potential contributors

Key sections:

  • What it does (one paragraph)
  • Installation instructions
  • Quick start example
  • Common use cases
  • Link to full documentation
  • How to contribute
  • License

2. API Documentation

Purpose: Help developers use your API correctly

Audience: Other developers integrating with your code

Key information:

  • Endpoint/function signature
  • Parameters and types
  • Return values
  • Error cases
  • Authentication requirements
  • Examples

3. Inline Comments

Purpose: Explain non-obvious decisions and complex logic

Audience: Future developers (including future you)

When to add:

  • Non-obvious decisions ("why" not "what")
  • Complex algorithms
  • Workarounds
  • Edge cases
  • Performance considerations
  • Security considerations

4. Technical Guides

Purpose: Teach how to accomplish specific tasks

Audience: Developers working with the system

Types:

  • How-to guides (step-by-step)
  • Architecture overviews (system structure)
  • Integration guides (connecting systems)
  • Troubleshooting guides (debugging help)

5. Architecture Decision Records (ADRs)

Purpose: Document important technical decisions and their rationale

Audience: Current and future team members

Key information:

  • Context (what problem exists)
  • Decision (what was chosen)
  • Alternatives considered
  • Consequences (trade-offs)

README Template

# Project Name

Brief description (one paragraph) of what this project does and why it exists.

## Features

- Feature 1
- Feature 2
- Feature 3

## Installation

npm install project-name


## Quick Start

import { Thing } from 'project-name'

const thing = new Thing() thing.doSomething() // Output: ...


## Usage

### Basic Usage

[Most common use case with complete example]

### Advanced Usage

[More complex scenarios]

## API Reference

### `functionName(param1, param2)`

Description of what the function does.

**Parameters:**

- `param1` (string, required): Description of parameter
- `param2` (number, optional): Description with default. Default: 42

**Returns:** `string` - Description of return value

**Throws:**

- `Error` - When parameter is invalid

**Example:**

const result = functionName('hello', 10) console.log(result) // "hello repeated 10 times"


## Configuration

[How to configure, if applicable]

## Troubleshooting

### Error: "Cannot find module"

**Cause:** Package not installed **Solution:**

npm install project-name


## Contributing

We welcome contributions! Please see [CONTRIBUTING.md](https://github.com/thebushidocollective/han/blob/HEAD/plugins/core/skills/documentation/CONTRIBUTING.md) for details.

## License

MIT License - see [LICENSE](https://github.com/thebushidocollective/han/blob/HEAD/plugins/core/skills/documentation/LICENSE) file for details.

API Documentation Template

REST API

## POST /api/users

Create a new user account.

### Authentication

Requires admin access token in Authorization header.

### Request

**Headers:**

{ "Content-Type": "application/json", "Authorization": "Bearer <admin_token>" }


**Body:**

{ "email": "user@example.com", "name": "John Doe", "role": "user" }


**Body Parameters:**

- `email` (string, required): Valid email address. Must be unique.
- `name` (string, required): User's full name. 2-100 characters.
- `role` (string, optional): User role. One of: "user", "admin". Default: "user".

### Response

**Success (201 Created):**

{ "id": "abc123", "email": "user@example.com", "name": "John Doe", "role": "user", "createdAt": "2024-01-01T00:00:00Z" }


**Error Responses:**

**400 Bad Request:**

{ "error": "Invalid email format", "field": "email" }


**401 Unauthorized:**

{ "error": "Authentication required" }


**403 Forbidden:**

{ "error": "Admin access required" }


**409 Conflict:**

{ "error": "Email already exists", "field": "email" }


**429 Too Many Requests:**

{ "error": "Rate limit exceeded. Try again in 60 seconds.", "retryAfter": 60 }


### Example

**cURL:**

curl -X POST https://api.example.com/users \ -H "Content-Type: application/json" \ -H "Authorization: Bearer abc123..." \ -d '{ "email": "user@example.com", "name": "John Doe", "role": "user" }'


**JavaScript:**

const response = await fetch('https://api.example.com/users', { method: 'POST', headers: { 'Content-Type': 'application/json', 'Authorization': 'Bearer abc123...' }, body: JSON.stringify({ email: 'user@example.com', name: 'John Doe', role: 'user' }) })

const user = await response.json() console.log(user.id) // "abc123"


### Rate Limiting

- 100 requests per minute per API key
- Returns 429 when exceeded
- Check `X-RateLimit-Remaining` header

Function/Method Documentation

/**
 * Calculate the total price including tax and shipping.
 *
 * Tax is calculated on the subtotal, not including shipping.
 * Free shipping is applied for orders over $50.
 *
 * @param items - Array of cart items with price and quantity
 * @param shippingAddress - Shipping address for tax calculation
 * @returns Object containing subtotal, tax, shipping, and total
 *
 * @throws {Error} If items array is empty
 * @throws {Error} If any item has invalid price or quantity
 *
 * @example
 * ```typescript
 * const total = calculateTotal([
 *   { price: 10, quantity: 2 },
 *   { price: 15, quantity: 1 }
 * ], { state: 'CA' })
 *
 * console.log(total)
 * // {
 * //   subtotal: 35,
 * //   tax: 2.80,
 * //   shipping: 0,
 * //   total: 37.80
 * // }
 * ```
 */
function calculateTotal(
  items: CartItem[],
  shippingAddress: Address
): OrderTotal {
  // Implementation...
}

Inline Comment Guidelines

When to Comment

✅ DO comment:

  • Why: Explains reasoning for non-obvious decisions
  • Trade-offs: Documents deliberate choices
  • Workarounds: Explains temporary solutions
  • Complex logic: Clarifies difficult algorithms
  • Edge cases: Documents special handling
  • Performance: Explains optimization choices
  • Security: Notes security considerations

❌ DON'T comment:

  • What: Code is self-explanatory
  • Redundant: Comment repeats code
  • Obvious: Anyone can see what it does

Good vs Bad Comments

// ❌ BAD: States the obvious
// Loop through users
for (const user of users) {
  // Print user name
  console.log(user.name)
}

// ✅ GOOD: Explains why
// Process in creation order to maintain referential integrity
// (newer records may reference older ones)
for (const user of users.sort((a, b) => a.createdAt - b.createdAt)) {
  processUser(user)
}

// ❌ BAD: Redundant
// Add 1 to counter
counter = counter + 1

// ✅ GOOD: Explains non-obvious decision
// +1 offset because API uses 1-based indexing (not 0-based)
const pageNumber = index + 1

// ❌ BAD: Explains what (obvious)
function calculateTotal(price, quantity) {
  // Multiply price by quantity
  return price * quantity
}

// ✅ GOOD: Explains why (non-obvious)
function calculateTotal(price, quantity) {
  // Use Number.toFixed(2) to prevent floating point errors
  // (0.1 + 0.2 !== 0.3 in JavaScript)
  return Number((price * quantity).toFixed(2))
}

// ❌ BAD: Commented-out code
function processOrder(order) {
  // const tax = order.subtotal * 0.08
  const tax = calculateTax(order)
  return order.subtotal + tax
}

// ✅ GOOD: Explains temporary workaround
function processOrder(order) {
  // TODO: Remove when new tax API is deployed (JIRA-123)
  // Using hardcoded rate temporarily during migration
  const tax = order.subtotal * 0.08
  return order.subtotal + tax
}

Comment Formats

Single-line comments:

// For brief explanations
const tax = subtotal * TAX_RATE  // Updated quarterly

Multi-line comments:

/*
 * For longer explanations that span multiple lines.
 * Use when context requires more detail.
 */

Doc comments (JSDoc/TSDoc):

/**
 * For API documentation.
 * Parsed by documentation generators.
 * @param name - Parameter description
 * @returns Return value description
 */

TODO comments:

// TODO(username): Description of what needs to be done
// FIXME: Description of what's broken and needs fixing
// HACK: Description of temporary workaround
// NOTE: Important information to highlight

Technical Guide Template

# How to [Task Name]

Brief description of what this guide teaches and who should use it.

## Prerequisites

- Requirement 1
- Requirement 2
- Knowledge/tools needed

## Overview

High-level explanation of the process or concept.

## Step-by-Step Instructions

### Step 1: [First Action]

Detailed description of what to do.

Command to run

command --option value


**Expected result:** What you should see.

**If something goes wrong:** How to troubleshoot.

### Step 2: [Next Action]

[Continue with clear steps...]

## Example

Complete working example showing the entire process.

// Full code example


## Common Issues

### Issue: [Problem Description]

**Symptom:** What you see when this happens. **Cause:** Why it happens. **Solution:** How to fix it.

## Related Documentation

- [Link to related guide]
- [Link to API docs]

## Next Steps

- [What to do after completing this guide]

Architecture Decision Record Template

# ADR-###: [Decision Title]

**Status:** Proposed | Accepted | Deprecated | Superseded
**Date:** YYYY-MM-DD
**Deciders:** [Names of decision makers]

## Context

What is the issue we're trying to solve? What factors are at play?
What constraints exist?

## Decision

What did we decide to do?

## Alternatives Considered

### Alternative 1: [Name]

**Description:** Brief explanation

**Pros:**
- Advantage 1
- Advantage 2

**Cons:**
- Disadvantage 1
- Disadvantage 2

**Why not chosen:** Explanation

### Alternative 2: [Name]

[Same structure...]

## Consequences

### Positive

- Benefit 1
- Benefit 2

### Negative

- Drawback 1
- Drawback 2

### Neutral

- Trade-off 1
- Trade-off 2

## Implementation Notes

Any specific guidance for implementing this decision.

## References

- [Related documentation]
- [Discussion links]
- [Research sources]

Documentation Best Practices

1. Write for Your Audience

Different audiences need different docs:

  • New users: Quick start, common use cases
  • API consumers: Complete reference, examples
  • Contributors: Architecture, dev setup
  • Maintainers: Design decisions, trade-offs

2. Show, Don't Just Tell

Bad:

The function accepts parameters and returns a result.

Good:

function calculateTotal(items, tax) {
  return items.reduce((sum, item) => sum + item.price, 0) * (1 + tax)
}

// Example usage:
const items = [{ price: 10 }, { price: 20 }]
const total = calculateTotal(items, 0.08) // 32.40

3. Keep Examples Runnable

Test your examples:

  • Copy/paste should work
  • Include all imports
  • Show complete code, not fragments
  • Verify examples actually run

4. Maintain Documentation

Documentation rots:

  • Update docs when code changes
  • Include doc updates in PRs
  • Review docs periodically
  • Mark outdated docs clearly

5. Use Clear Language

Avoid jargon:

  • Bad: "Leverages polymorphic memoization"
  • Good: "Caches results based on input type"

Be specific:

  • Bad: "It's fast"
  • Good: "Responds in < 100ms for 95% of requests"

Use active voice:

  • Bad: "The request is processed by the server"
  • Good: "The server processes the request"

Documentation Checklist

README

  • Clear description of what project does
  • Installation instructions
  • Quick start example
  • Link to full documentation
  • How to contribute
  • License information

API Docs

  • All public APIs documented
  • Parameters described with types
  • Return values described
  • Error cases documented
  • Examples provided
  • Authentication requirements stated

Inline Comments

  • Complex logic explained
  • Non-obvious decisions documented
  • Workarounds noted
  • No redundant comments
  • No commented-out code
  • TODOs have owner and context

Technical Guides

  • Prerequisites listed
  • Steps are clear and ordered
  • Examples are complete and runnable
  • Common issues addressed
  • Related docs linked

Integration with Other Skills

  • Use explainer skill for creating clear explanations
  • Use proof-of-work skill to verify examples actually work
  • Reference architecture-design skill for architecture documentation
  • Reference technical-planning skill for implementation guides

Common Documentation Mistakes

❌ No Examples

Bad: "Use the function to process data" Good: "Use processData(items) to validate and format user input"

❌ Outdated Information

Bad: Docs say version 1.0, code is version 3.0 Good: Update docs with code changes

❌ Assuming Knowledge

Bad: "Simply configure the flux capacitor" Good: "Add FLUX_CAPACITOR=true to your.env file"

❌ Incomplete Examples

Bad:

const result = doSomething()

Good:

import { doSomething } from 'my-package'

const result = doSomething({
  option1: 'value',
  option2: 42
})
console.log(result) // Expected output

❌ Too Much Documentation

Bad: Document every single line Good: Document the "why" and non-obvious parts

Remember

  1. Write for the reader - What do they need to know?
  2. Show with examples - Working code is clearer than prose
  3. Keep it updated - Stale docs are worse than no docs
  4. Test examples - Verify they actually work
  5. Explain "why" - The "what" is in the code

Good documentation helps people use your code correctly and efficiently.

平台分布

Claude Code

30.06%
按下载量换算20

OpenCode

23.1%
按下载量换算15

Codex

16.83%
按下载量换算11

Antigravity

13.32%
按下载量换算9

windsurf

7.7%
按下载量换算5

Gemini CLI

3.66%
按下载量换算2

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。

继续浏览同类 Skills