Token导航 LogoToken导航

Agent Skill · 研究检索

workspace-documentation(工作区文档)

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。

workspace-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,权限为只读;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 36,780。

总安装 1,860周安装 76GitHub Stars 5下载 602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:workspace-documentation(工作区文档)
来源仓库:https://github.com/qodex-ai/ai-agent-skills
仓库路径:skills/workspace-documentation
安装命令:
npx skills add https://github.com/qodex-ai/ai-agent-skills --skill workspace-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/qodex-ai/ai-agent-skills --skill workspace-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 和内容稿件的整理与改写。适合提炼结构、统一术语或检查链接有效性。应保留项目已有事实,避免写成确定结论。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。涉及对外文案时需控制语气,避免过度营销。

workspace-documentation 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/workspace-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/qodex-ai/ai-agent-skills --skill workspace-doc…;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

workspace-documentation 怎么安装?

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

workspace-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。

workspace-documentation 需要什么权限?

权限级别是只读。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。

workspace-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。

workspace-documentation 用的人多吗?

公开数据是累计安装 1,860 次,近一周安装 76 次,下载 602 次,GitHub 5 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。

workspace-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

  • 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
workspace-documentation

工作区文档 · 本页

1,8605只读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

Knowledge Capture

Transforms conversations, discussions, and insights into structured documentation in your Notion workspace. Captures knowledge from chat context, formats it appropriately, and saves it to the right location with proper organization and linking.

Quick Start

When asked to save information to Notion:

  1. Extract content: Identify key information from conversation context
  2. Structure information: Organize into appropriate documentation format
  3. Determine location: Use Notion:notion-search to find appropriate wiki page/database
  4. Create page: Use Notion:notion-create-pages to save content
  5. Make discoverable: Link from relevant hub pages, add to databases, or update wiki navigation so others can find it

Knowledge Capture Workflow

Step 1: Identify content to capture

From conversation context, extract:
- Key concepts and definitions
- Decisions made and rationale
- How-to information and procedures
- Important insights or learnings
- Q&A pairs
- Examples and use cases

Step 2: Determine content type

Classify the knowledge:
- Concept/Definition
- How-to Guide
- Decision Record
- FAQ Entry
- Meeting Summary
- Learning/Post-mortem
- Reference Documentation

Step 3: Structure the content

Format appropriately based on content type:
- Use templates for consistency
- Add clear headings and sections
- Include examples where helpful
- Add relevant metadata
- Link to related pages

Step 4: Determine destination

Where to save:
- Wiki page (general knowledge base)
- Specific project page (project-specific knowledge)
- Documentation database (structured docs)
- FAQ database (questions and answers)
- Decision log (architecture/product decisions)
- Team wiki (team-specific knowledge)

Step 5: Create the page

Use Notion:notion-create-pages:
- Set appropriate title
- Use structured content from template
- Set properties if in database
- Add tags/categories
- Link to related pages

Step 6: Make content discoverable

Link the new page so others can find it:

1. Update hub/index pages:
   - Add link to wiki table of contents page
   - Add link from relevant project page
   - Add link from category/topic page (e.g., "Engineering Docs")

2. If page is in a database:
   - Set appropriate tags/categories
   - Set status (e.g., "Published")
   - Add to relevant views

3. Optionally update parent page:
   - If saved under a project, add to project's "Documentation" section
   - If in team wiki, ensure it's linked from team homepage

Example:
Notion:notion-update-page
page_id: "team-wiki-homepage-id"
command: "insert_content_after"
selection_with_ellipsis: "## How-To Guides..."
new_str: "- <mention-page url='...'>How to Deploy to Production</mention-page>"

This step ensures the knowledge doesn't become "orphaned" - it's properly connected to your workspace's navigation structure.

Content Types

Choose appropriate structure based on content:

Concept: Overview → Definition → Characteristics → Examples → Use Cases → Related How-To: Overview → Prerequisites → Steps (numbered) → Verification → Troubleshooting → Related Decision: Context → Decision → Rationale → Options Considered → Consequences → Implementation FAQ: Short Answer → Detailed Explanation → Examples → When to Use → Related Questions Learning: What Happened → What Went Well → What Didn't → Root Causes → Learnings → Actions

Destination Patterns

General Wiki: Standalone page → add to index → tag → link from related pages

Project Wiki: Child of project page → link from project overview → tag with project name

Documentation Database: Use properties (Title, Type, Category, Tags, Last Updated, Owner)

Decision Log Database: Use properties (Decision, Date, Status, Domain, Deciders, Impact)

FAQ Database: Use properties (Question, Category, Tags, Last Reviewed, Useful Count)

See reference/database-best-practices.md for database selection guide and individual schema files.

Content Extraction from Conversations

Chat Discussion: Key points, conclusions, resources, action items, Q&A

Problem-Solving: Problem statement, approaches tried, solution, why it worked, future considerations

Knowledge Sharing: Concept explained, examples, best practices, common pitfalls, resources

Decision Discussion: Question, options, trade-offs, decision, rationale, next steps

Formatting Best Practices

Structure: Use # (title), ## (sections), ### (subsections) consistently

Writing: Start with overview, use bullets, keep paragraphs short, add examples

Linking: Link related pages, mention people, reference resources, create bidirectional links

Metadata: Include date, author, tags, status

Searchability: Clear titles, natural keywords, common search tags, image alt-text

Indexing and Organization

Wiki Index: Organize by sections (Getting Started, How-To Guides, Reference, FAQs, Decisions) with page links

Category Pages: Create landing pages with overview, doc links, and recent updates

Tagging Strategy: Use consistent tags for technology/tools, topics, audience, and status

Update Management

Create New: Content is substantive (>2 paragraphs), will be referenced multiple times, part of knowledge base, needs independent discovery

Update Existing: Adding to existing topic, correcting info, expanding concept, updating for changes

Versioning: Add update history section for significant changes (date, author, what changed, why)

Best Practices

  1. Capture promptly: Document while context is fresh
  2. Structure consistently: Use templates for similar content
  3. Link extensively: Connect related knowledge
  4. Write for discovery: Use searchable titles and tags
  5. Include context: Why this matters, when to use
  6. Add examples: Concrete examples aid understanding
  7. Maintain: Review and update periodically
  8. Get feedback: Ask if documentation is helpful

Advanced Features

Documentation databases: See reference/database-best-practices.md for database schema patterns.

Common Issues

"Not sure where to save": Default to general wiki, can move later "Content is fragmentary": Group related fragments into cohesive doc "Already exists": Search first, update existing if appropriate "Too informal": Clean up language while preserving insights

Examples

See examples/ for complete workflows:

平台分布

Claude Code

30.9%
按下载量换算186

trae

22.9%
按下载量换算138

Codex

18.08%
按下载量换算109

Antigravity

11.75%
按下载量换算71

windsurf

7.56%
按下载量换算46

github-copilot

3.35%
按下载量换算20

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。

继续浏览同类 Skills