Token导航 LogoToken导航

Agent Skill · 研究检索

document-writing-skills(文档写作技巧)

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。

document-writing-skills 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,权限为需要联网;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 53,488。

总安装 533周安装 22GitHub Stars 4下载 174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:document-writing-skills(文档写作技巧)
来源仓库:https://github.com/kimasplund/claude_cognitive_reasoning
仓库路径:skills/document-writing-skills
安装命令:
npx skills add https://github.com/kimasplund/claude_cognitive_reasoning --skill document-writing-skills
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/kimasplund/claude_cognitive_reasoning --skill document-writing-skills

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 和说明材料的整理与改写,适合提炼结构、补齐章节或统一术语。适用于需要生成专业文档、保持内容一致性和清晰度的场景,如项目说明、技术文档或对外文案。通过模板和规范确保文档质量,使用时需保留事实信息,避免将未确认内容写成确定结论。

安装方式:通过 GitHub 仓库使用 npx 命令添加,支持 Codex、Claude、Cursor 等宿主环境。涉及对外文案时需注意语气控制,避免过度营销或夸大能力描述。

来源仓库内路径:skills/document-writing-skills

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/kimasplund/claude_cognitive_reasoning --skill…;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

    它需要需要联网,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

document-writing-skills 怎么安装?

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

document-writing-skills 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。

document-writing-skills 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

document-writing-skills 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。

document-writing-skills 用的人多吗?

公开数据是累计安装 533 次,近一周安装 22 次,下载 174 次,GitHub 4 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。

document-writing-skills 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

  • 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

  • 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
document-writing-skills

文档写作技巧 · 本页

5334需要联网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

Document Writing Skills

Purpose: This skill provides comprehensive document writing patterns, templates, and best practices that agents can apply when generating professional documentation, reports, contracts, guides, and technical writing across various domains.

When to Use Document Writing Skills

Use this skill when:

  • Creating API documentation (REST/GraphQL/RPC)
  • Writing technical documentation or user guides
  • Generating research reports or incident reports
  • Drafting architecture decision records (ADRs)
  • Creating changelogs and release notes
  • Writing legal memoranda or contracts
  • Producing test reports or security assessments
  • Authoring product requirement documents (PRDs)

Core Writing Principles

1. Clarity and Conciseness

Guidelines:

  • Use active voice: "The system processes requests" (not "Requests are processed by the system")
  • Use present tense: "The function returns" (not "The function will return")
  • Be specific: "Response time: 200ms" (not "Response time is fast")
  • Avoid jargon unless domain-appropriate
  • Use short sentences (15-20 words maximum)
  • Break complex ideas into numbered steps

Example - Before and After:

❌ Before: "It should be noted that the API endpoint might be utilized for the purpose of retrieving user data."
✅ After: "Use this endpoint to retrieve user data."

2. Progressive Disclosure Structure

Organize documents from high-level to detailed:

  1. Summary/Overview - What and why (2-3 sentences)
  2. Key Concepts - Essential understanding
  3. Details - Deep-dive information
  4. Reference - Complete specifications

Template:

# Document Title

## Summary
[2-3 sentences: What this is and why it matters]

## Quick Start
[Minimal steps to get started]

## Concepts
[Essential understanding]

## Detailed Guide
[In-depth information]

## Reference
[Complete specifications, API details, etc.]

3. Consistency Standards

Maintain consistency in:

  • Terminology (create glossary for domain terms)
  • Code formatting (use syntax highlighting)
  • Section structure (follow templates)
  • Date formats (ISO 8601: YYYY-MM-DD)
  • Version numbers (Semantic Versioning: MAJOR.MINOR.PATCH)

Document Type Templates

API Documentation

Structure:

# API Endpoint: [Method] /path/to/endpoint

## Overview
[Brief description of what this endpoint does]

## Authentication
[Required authentication method]

## Request

**Method**: [GET/POST/PUT/DELETE]
**URL**: `/api/v1/endpoint`

**Headers**:
| Header | Value | Required |
|--------|-------|----------|
| Authorization | Bearer {token} | Yes |
| Content-Type | application/json | Yes |

**Parameters**:
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| user_id | integer | Yes | Unique user identifier |
| limit | integer | No | Max results (default: 10) |

**Request Body** (JSON):

{ "field1": "value", "field2": 123 }


## Response

**Success Response** (200 OK):

{ "status": "success", "data": { "id": 12345, "name": "Example" } }


**Error Responses**:

| Status Code | Description | Response Body |
| --- | --- | --- |
| 400 | Bad Request | `{"error": "Invalid parameters"}` |
| 401 | Unauthorized | `{"error": "Authentication required"}` |
| 404 | Not Found | `{"error": "Resource not found"}` |
| 500 | Server Error | `{"error": "Internal server error"}` |

## Example Usage

**cURL**:

curl -X POST https://api.example.com/v1/endpoint \ -H "Authorization: Bearer YOUR_TOKEN" \ -H "Content-Type: application/json" \ -d '{"field1": "value"}'


**Python**:

import requests

response = requests.post( "https://api.example.com/v1/endpoint", headers={"Authorization": "Bearer YOUR_TOKEN"}, json={"field1": "value"} )


## Rate Limiting

[Rate limit details if applicable]

## Notes

[Additional considerations, edge cases, deprecation warnings]

Technical Report

Structure:

# [Report Title]

**Author**: [Name/Role]
**Date**: [YYYY-MM-DD]
**Version**: [X.Y]
**Status**: [Draft/Final/Approved]

## Executive Summary
[2-3 paragraphs: Key findings, recommendations, impact]

## Background
[Context: Why this report exists, what problem it addresses]

## Methodology
[How the analysis was conducted, tools used, data sources]

## Findings

### Finding 1: [Title]
**Impact**: [High/Medium/Low]
**Evidence**: [Data, observations, metrics]
**Analysis**: [What this means]

### Finding 2: [Title]
[Same structure...]

## Recommendations

### Recommendation 1: [Action]
**Priority**: [High/Medium/Low]
**Effort**: [High/Medium/Low]
**Timeline**: [Timeframe]
**Rationale**: [Why this recommendation]

## Conclusion
[Summary of key points and next steps]

## Appendix
[Supporting data, detailed tables, raw results]

Architecture Decision Record (ADR)

Structure:

# ADR-[Number]: [Title]

**Status**: [Proposed/Accepted/Deprecated/Superseded]
**Date**: [YYYY-MM-DD]
**Deciders**: [Names/Roles]

## Context
[What is the issue we're addressing?]

## Decision
[What decision have we made?]

## Rationale
[Why did we make this decision?]

### Options Considered

#### Option 1: [Name]
**Pros**:
- [Advantage 1]
- [Advantage 2]

**Cons**:
- [Disadvantage 1]
- [Disadvantage 2]

#### Option 2: [Name]
[Same structure...]

### Decision Criteria
- [Criterion 1: Performance]
- [Criterion 2: Maintainability]
- [Criterion 3: Cost]

## Consequences

### Positive
- [Benefit 1]
- [Benefit 2]

### Negative
- [Trade-off 1]
- [Trade-off 2]

### Neutral
- [Impact 1]

## Implementation Notes
[Guidance for implementing this decision]

## Related Decisions
- [ADR-XXX: Related decision]

## References
- [Documentation links, research papers, discussions]

User Guide

Structure:

# [Product/Feature] User Guide

## Overview
[What this is and what it helps users accomplish]

## Getting Started

### Prerequisites
- [Requirement 1]
- [Requirement 2]

### Installation
1. [Step 1 with validation]

command-to-run


**Expected output**: [What success looks like]

1. [Step 2] [Validation checkpoint]

## Tutorials

### Tutorial 1: [Basic Task]

**Goal**: [What you'll accomplish] **Time**: [Estimated duration]

**Steps**:

1. [Action]
  - **Why**: [Rationale]
  - **Validation**: [How to verify success]
2. [Next action] [Same structure...]

**Result**: [What you've achieved]

### Tutorial 2: [Advanced Task]

[Same structure...]

## Reference

### Feature 1

**Description**: [What it does] **Usage**: [How to use it] **Parameters**: [Options available] **Examples**: [Common use cases]

## Troubleshooting

### Problem: [Common issue]

**Symptoms**: [How to identify this problem] **Cause**: [Why this happens] **Solution**: [Step-by-step fix]

### Problem: [Another issue]

[Same structure...]

## FAQ

**Q: [Common question]** A: [Clear answer]

## Additional Resources

- [Link to API docs]
- [Link to community forum]

Changelog

Format: Keep a Changelog (https://keepachangelog.com/en/1.0.0/)

Structure:

# Changelog

All notable changes to this project will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

## [Unreleased]

### Added
- [New feature description]

### Changed
- [Change to existing functionality]

### Deprecated
- [Soon-to-be removed feature]

### Removed
- [Removed feature]

### Fixed
- [Bug fix]

### Security
- [Vulnerability fix]

## [1.2.0] - 2025-11-08

### Added
- User authentication with JWT tokens
- Password reset functionality
- Email verification

### Changed
- Improved API response times by 40%
- Updated database schema for better performance

### Fixed
- Issue #123: Login timeout on slow connections
- Memory leak in background worker

## [1.1.0] - 2025-10-15

[Previous version changes...]

## [1.0.0] - 2025-09-01

### Added
- Initial release
- User registration and login
- Basic profile management

[Unreleased]: https://github.com/org/repo/compare/v1.2.0...HEAD
[1.2.0]: https://github.com/org/repo/compare/v1.1.0...v1.2.0
[1.1.0]: https://github.com/org/repo/compare/v1.0.0...v1.1.0
[1.0.0]: https://github.com/org/repo/releases/tag/v1.0.0

Breaking Changes Callout:

## [2.0.0] - 2025-11-08

### ⚠️ BREAKING CHANGES
- **Authentication**: Changed from session-based to JWT tokens
  - **Migration**: See [Migration Guide](./MIGRATION.md)
  - **Impact**: All clients must update authentication flow
- **API Endpoints**: Removed deprecated `/v1/users` endpoint
  - **Replacement**: Use `/v2/users` instead

Visual Elements

Code Examples

Always include:

  • Syntax highlighting (specify language)
  • Complete working examples (not fragments)
  • Expected output
  • Error handling

Template:

Complete example with imports

import requests

def get_user(user_id): """Retrieve user by ID.""" try: response = requests.get(f"https://api.example.com/users/{user_id}") response.raise_for_status() return response.json() except requests.exceptions.RequestException as e: print(f"Error: {e}") return None

Usage

user = get_user(123) print(user)


**Expected output**:

{ "id": 123, "name": "John Doe", "email": "john@example.com" }

Diagrams

When to include:

  • Architecture overviews
  • Data flow diagrams
  • State machines
  • Process workflows

Formats:

  • Mermaid (for text-based diagrams in Markdown)
  • ASCII art (for simple diagrams)
  • Image files (for complex visuals)

Mermaid Example:

sequenceDiagram participant Client participant API participant Database

Client->>API: POST /users API->>Database: INSERT user Database-->>API: User created API-->>Client: 201 Created

Tables

Use tables for:

  • Parameter specifications
  • Configuration options
  • Comparison matrices
  • Error code listings

Best practices:

  • Include headers
  • Align columns consistently
  • Use separators for readability
  • Keep cell content concise

Domain-Specific Patterns

Legal Documents

Key requirements:

  • Formal tone and precise language
  • Numbered sections and subsections
  • Citation formatting (jurisdiction-specific)
  • Definitions section
  • Signature blocks

Reference: See references/legal-document-patterns.md for:

  • Contract templates
  • Memorandum structure
  • Citation formats (Finlex, EUR-Lex, Bluebook, etc.)
  • Legal writing conventions

Security Reports

Structure:

  • Executive summary with risk scores
  • CVSS scoring for vulnerabilities
  • STRIDE/OWASP categorization
  • Remediation guidance with timelines
  • Evidence and proof-of-concept

Reference: See references/security-report-patterns.md for:

  • Vulnerability report template
  • Threat modeling documentation
  • Penetration test report structure
  • Security assessment framework

Product Requirements Document (PRD)

Structure:

  • Product overview and goals
  • User personas and use cases
  • Functional requirements
  • Non-functional requirements
  • Success metrics
  • Release criteria

Reference: See references/prd-patterns.md for:

  • Complete PRD template
  • Feature prioritization frameworks (RICE, MoSCoW)
  • User story formats
  • Acceptance criteria patterns

Quality Checklist

Before finalizing any document, verify:

Content Quality:

  • Purpose is clear in first paragraph
  • All technical terms are defined or linked
  • Examples are complete and tested
  • No assumptions about reader knowledge
  • Consistent terminology throughout

Structure Quality:

  • Logical flow from high-level to detailed
  • Table of contents for long documents (>5 sections)
  • Clear section headings
  • Appropriate use of lists vs paragraphs
  • Progressive disclosure applied

Technical Quality:

  • Code examples use syntax highlighting
  • All code has been validated
  • API endpoints include full URL
  • Error codes are documented
  • Links are valid and versioned

Accessibility:

  • Headings use proper hierarchy (H1 > H2 > H3)
  • Tables have descriptive headers
  • Images have alt text
  • Color is not the only indicator
  • Contrast ratios meet WCAG AA standards

Metadata:

  • Document has creation date (ISO 8601)
  • Author/owner is identified
  • Version number is specified
  • Last updated date is current
  • Status is indicated (Draft/Final/Deprecated)

Advanced Patterns

For detailed guidance on specialized documentation types, see:

  • Legal Document Patterns: references/legal-document-patterns.md

- Contract templates (employment, service agreements, NDAs) - Legal memorandum structure - Citation formats by jurisdiction - Finnish/EU legal writing conventions

  • Security Report Patterns: references/security-report-patterns.md

- Vulnerability disclosure templates - Threat modeling documentation - Penetration test report structure - CVSS scoring guidance

  • Product Document Patterns: references/prd-patterns.md

- Product requirement documents (PRDs) - Feature specification templates - User story and acceptance criteria formats - Prioritization frameworks (RICE, MoSCoW, Kano)

  • Technical Writing Best Practices: references/technical-writing-guide.md

- Style guide recommendations - Grammar and punctuation rules - Localization considerations - Inclusive language guidelines


Summary

This skill provides comprehensive patterns and templates for creating professional documentation across multiple domains. Apply these patterns to ensure consistency, clarity, and quality in all written deliverables.

Key Principles:

  1. Clarity: Active voice, present tense, specific language
  2. Structure: Progressive disclosure from high-level to detailed
  3. Consistency: Templates, terminology, formatting
  4. Completeness: Examples, error handling, validation
  5. Quality: Verified code, tested links, proper metadata

平台分布

Antigravity

25.27%
按下载量换算44

Claude Code

22.58%
按下载量换算39

Gemini CLI

18.49%
按下载量换算32

OpenCode

13.49%
按下载量换算23

windsurf

7.53%
按下载量换算13

github-copilot

3.47%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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