Token导航 LogoToken导航

Agent Skill · 研究检索

diagram(diagram 图表绘制)

diagram 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

diagram 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 51,787。

总安装 574周安装 23GitHub Stars 23下载 186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:diagram(diagram 图表绘制)
来源仓库:https://github.com/johnlindquist/claude
仓库路径:skills/diagram
安装命令:
npx skills add https://github.com/johnlindquist/claude --skill diagram
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/johnlindquist/claude --skill diagram

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。

diagram 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位结果。适用于需要基于任务场景或来源线索进行信息筛选的研究检索场景。通过 npx skills add 命令从指定 GitHub 仓库安装并使用该技能。

安装前需确认权限范围、维护状态,以及是否会触发联网或文件读写操作。建议结合原始 README 和仓库内容进一步核验具体用法和功能细节。

来源仓库内路径:skills/diagram

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/johnlindquist/claude --skill diagram;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

diagram 怎么安装?

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

diagram 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。

diagram 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。

diagram 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。

diagram 用的人多吗?

公开数据是累计安装 574 次,近一周安装 23 次,下载 186 次,GitHub 23 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。

diagram 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
diagram

diagram 图表绘制 · 本页

57423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

Diagram Driven Development (DDD) Skill

Maintain the ai/diagrams directory as the single source of truth for system understanding. All diagrams follow DDD principles, connecting Front-Stage (user experience) to Back-Stage (technical implementation) with clear impact annotations.

Capabilities

  1. Create Diagrams - Generate new diagrams for features, architectures, journeys, tests, and refactorings
  2. Update Diagrams - Synchronize existing diagrams with code changes
  3. Audit Diagrams - Identify outdated, missing, or low-quality diagrams
  4. Organize Diagrams - Maintain consistent structure and naming conventions
  5. Index Management - Keep README.md index up-to-date with all diagrams
  6. Quality Validation - Ensure all diagrams follow DDD principles

Quick Reference

For detailed instructions on each operation, see:

Directory Structure

ai/diagrams/
├── README.md                    # Index of all diagrams
├── features/                    # Feature-specific diagrams
├── architecture/                # System architecture diagrams
├── journeys/                    # User journey diagrams
├── tests/                       # Test coverage diagrams
└── refactoring/                 # Before/After improvement diagrams

Common Workflows

Initial Setup Workflow

  1. User starts new project or adds DDD to existing project
  2. Create ai/diagrams/ directory structure
  3. Generate initial system architecture diagram
  4. Create README.md index
  5. Document key user journeys

New Feature Workflow

  1. User requests new feature
  2. Create feature diagram showing user value
  3. Connect Front-Stage (UX) to Back-Stage (implementation)
  4. Document related files and components
  5. Update README.md index

Code Change Workflow

  1. Code is modified (new features, refactoring, etc.)
  2. Identify affected diagrams
  3. Update diagrams to reflect changes
  4. Update "Last Updated" dates
  5. Add change history entries

Audit Workflow

  1. User requests diagram audit
  2. Scan all diagrams in ai/diagrams/
  3. Check for outdated diagrams (compare dates with git)
  4. Identify missing diagrams (features without diagrams)
  5. Validate DDD quality (Front-Stage/Back-Stage, impact annotations)
  6. Report findings and recommendations

Refactoring Documentation Workflow

  1. User plans code refactoring
  2. Create "Before" diagram showing current state
  3. Create "After" diagram showing improved state (highlight changes in #90EE90)
  4. Add impact annotations explaining user benefits
  5. Store in refactoring/ directory

Critical Instructions

REQUIRED: Before performing ANY diagram operations, you MUST load the relevant reference file(s) using the Read tool. These references contain essential DDD principles, quality standards, and operational procedures that are NOT included in this overview.

When the user asks to work with diagrams:

  1. Identify the operation they want to perform (create, update, audit, organize)
  2. MANDATORY: Load the relevant reference file(s) using the Read tool BEFORE executing any operations:

- Creating diagrams → Read references/CREATE.md AND references/DDD_PRINCIPLES.md FIRST - Updating diagrams → Read references/UPDATE.md AND references/DDD_PRINCIPLES.md FIRST - Auditing diagrams → Read references/AUDIT.md FIRST - Organizing/restructuring → Read references/ORGANIZE.md FIRST - Understanding DDD → Read references/DDD_PRINCIPLES.md FIRST - Mermaid syntax help → Read references/MERMAID_GUIDE.md FIRST

  1. Execute diagram operations following the exact patterns and quality standards from the loaded references
  2. Validate quality using DDD principles checklist
  3. Update index in README.md to reflect changes
  4. Confirm actions and show diagram preview when possible

DO NOT attempt to create or modify diagrams without first loading and reading the relevant reference documentation, especially DDD_PRINCIPLES.md.

DDD Core Principles (Brief)

Every diagram MUST include:

  • ✅ Front-Stage (user experience) AND Back-Stage (implementation)
  • ✅ Impact Annotations explaining user value of technical components
  • ✅ User Actions as entry/exit points
  • ✅ Error Paths and recovery options
  • ✅ Related Files documentation
  • ❌ NO custom fill colors (except #90EE90 for Before/After changes)
  • ❌ NO purely technical diagrams without user context

Naming Conventions

File Names

  • Descriptive lowercase with hyphens
  • Include diagram type prefix
  • Format: {type}-{descriptive-name}.md

Examples:

  • feature-user-checkout-flow.md
  • sequence-authentication-journey.md
  • arch-system-overview.md
  • flow-payment-processing.md

Type Prefixes

  • feature- - Feature-specific diagrams
  • sequence- - Sequence/journey diagrams
  • arch- - Architecture diagrams
  • flow- - Flow/process diagrams
  • test- - Test coverage diagrams

Diagram File Structure

# [Diagram Title]

**Type:** [Feature Diagram | Sequence Diagram | Architecture Diagram | etc.]
**Last Updated:** [YYYY-MM-DD]
**Related Files:**
- `path/to/implementation.ts`
- `path/to/component.tsx`

## Purpose

[1-2 sentence description of what user value this diagram illustrates]

## Diagram

\`\`\`mermaid
[Mermaid diagram code following DDD principles]
\`\`\`

## Key Insights

- [User impact point 1]
- [User impact point 2]
- [Technical enabler point 1]

## Change History

- **YYYY-MM-DD:** [Description of change]

Quality Checklist

Before storing any diagram, verify:

  • Shows both Front-Stage (user experience) AND Back-Stage (implementation)
  • Impact annotations explain user value
  • User actions are clearly visible
  • Error paths shown
  • NO custom fill colors (except #90EE90 for changes)
  • Related code files documented
  • Last updated date is current
  • Key insights explain user impact
  • Mermaid syntax is valid

Best Practices

  1. Keep diagrams synchronized - Outdated diagrams are worse than no diagrams
  2. Follow DDD principles - Every diagram connects user value to implementation
  3. Use subdirectories - Organize by type to prevent chaos
  4. Maintain the index - README.md is the entry point
  5. Document changes - Update change history when modifying
  6. Validate quality - Run through DDD checklist before saving
  7. Reference code files - Link diagrams to actual implementation
  8. Show error paths - Don't just show happy paths
  9. Use consistent naming - Predictable names enable navigation
  10. Update after code changes - Diagrams must reflect current state

Integration with Other Skills

  • review - Reference diagrams during code reviews to explain impact
  • github - Link diagrams in issue descriptions for context
  • chrome-devtools - Use diagrams to plan testing flows

Examples

Create feature diagram

User: "Create a diagram for the new notification system"
Agent:
1. Reads references/CREATE.md and references/DDD_PRINCIPLES.md
2. Analyzes notification feature code
3. Creates feature-notification-system.md in features/
4. Includes user journey and technical implementation
5. Adds impact annotations
6. Updates README.md index

Update after refactoring

User: "We just refactored the auth flow, update the diagram"
Agent:
1. Reads references/UPDATE.md
2. Finds sequence-authentication-journey.md
3. Compares with new code
4. Updates diagram with changes
5. Updates "Last Updated" date
6. Adds change history entry

Audit all diagrams

User: "Audit our diagrams"
Agent:
1. Reads references/AUDIT.md and references/DDD_PRINCIPLES.md
2. Scans ai/diagrams/ directory
3. Checks each diagram against DDD checklist
4. Compares diagram dates with git history
5. Identifies missing diagrams
6. Reports findings with recommendations

Critical Rules

  1. Diagrams MUST stay synchronized with code - Check git history vs diagram dates
  2. Every diagram MUST follow DDD principles - No purely technical diagrams
  3. Organization is critical - Use subdirectories consistently
  4. Index MUST be maintained - README.md reflects all diagrams
  5. File naming MUST be consistent - Follow type-name pattern
  6. Quality over quantity - Better to have 5 great diagrams than 20 poor ones
  7. User value is paramount - Every technical detail must connect to user impact
  8. Always load references first - DDD principles are not negotiable

Workflow Integration

This skill integrates with development workflow:

  1. Before Code Changes - Review existing diagrams to understand system
  2. During Planning - Create proposal diagrams showing planned changes
  3. During Implementation - Reference diagrams to maintain alignment
  4. After Implementation - Update diagrams to reflect changes
  5. During Review - Use diagrams to explain impact and context
  6. During Onboarding - Diagrams serve as documentation for new team members

平台分布

Claude Code

26.95%
按下载量换算50

OpenCode

25.03%
按下载量换算47

Antigravity

16.49%
按下载量换算31

Gemini CLI

14.34%
按下载量换算27

windsurf

8.45%
按下载量换算16

trae

3.66%
按下载量换算7

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