本文へ移動
cccskills
無料GitHub で公開

diagramming-expert

Master of text-based visual communication using ASCII art, Unicode box-drawing, and structured diagram notation. Creates clear, maintainable diagrams for systems, processes, hierarchies, relationships, and psychological structures. Proactively generates diagrams to enhance understanding. Activate on visualization needs, system architecture, process flows, psychological mapping, or when complex concepts would benefit from visual representation. NOT for photo editing, vector graphics, or GUI-based design tools.

インストール方法を見る

含まれるファイル(6)

  • SKILL.md12.2 KB
  • references/diagram-types.md27.4 KB
  • references/INDEX.md505 B
  • references/jungian-diagrams.md55.5 KB
  • references/patterns-library.md33.5 KB
  • references/unicode-characters.md6.8 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Diagramming Expert

Master of text-based visual communication. Creates clear, maintainable diagrams that reduce cognitive load and enhance understanding.

Decision Points

Diagram Type Selection Tree

Problem Size Analysis:
├── Small (≤5 elements)
│   ├── If showing relationships → Simple network diagram
│   ├── If showing sequence → Linear flowchart
│   └── If showing hierarchy → Tree diagram
├── Medium (6-15 elements)
│   ├── If showing process → Multi-lane flowchart with decision points
│   ├── If showing architecture → Layered architecture diagram
│   └── If showing relationships → Clustered network with grouping
└── Large (>15 elements)
    ├── If stakeholders = technical → Break into multiple detailed diagrams
    ├── If stakeholders = business → Create overview + drill-down diagrams
    └── If medium = presentation → Create progression slides

Complexity Threshold Heuristics:
- If >10 arrows cross each other → Break into multiple diagrams
- If labels don't fit in 80-character width → Simplify or split
- If >7 decision points in one flow → Create decision matrix instead
- If >5 hierarchical levels → Consider nested zoom-in approach

Layout Decision Matrix

Content Type:
├── Static structure
│   ├── Top-down hierarchy → Use tree layout
│   ├── Peer relationships → Use horizontal layout
│   └── Layered system → Use vertical stack
├── Dynamic process
│   ├── Linear sequence → Left-to-right flow
│   ├── Branching logic → Decision tree format
│   └── Cyclical process → Circular or loop layout
└── Mixed content
    ├── Structure + process → Swimlane diagram
    ├── Multiple viewpoints → Side-by-side comparison
    └── Temporal changes → Before/after layout

Annotation Strategy Decision

Information Density:
├── High detail required
│   ├── Use numbered callouts with legend below
│   ├── Create detail boxes connected by dotted lines
│   └── Split into overview + detail diagrams
├── Medium detail
│   ├── Inline labels within boxes
│   ├── Short descriptive text near elements
│   └── Color coding with simple legend
└── Low detail (overview)
    ├── Single-word labels only
    ├── Focus on structure, not content
    └── Use size/weight to show importance

Failure Modes

1. Spaghetti Syndrome

Symptoms: Crossed arrows everywhere, unclear information flow, reader can't follow path
Detection Rule: If you count >5 line crossings or can't trace any path without confusion
Fix: Reorganize layout to minimize crossings, use hierarchical structure, break into multiple focused diagrams

2. Information Overload

Symptoms: Everything in one diagram, tiny unreadable text, cramped elements
Detection Rule: If diagram requires >120 character width or elements are <3 characters apart
Fix: Apply "One Main Idea" rule - create overview + detail breakouts, use progressive disclosure

3. Chartjunk Bloat

Symptoms: Excessive decoration, multiple border styles, unnecessary graphical elements
Detection Rule: If more than 2 different line styles or decorative elements not carrying information
Fix: Apply Tufte's data-ink ratio - remove all non-essential visual elements, use consistent minimal styling

4. Unlabeled Mystery

Symptoms: Boxes without labels, arrows without direction indicators, unclear relationships
Detection Rule: If any element requires more than 5 seconds to understand its purpose
Fix: Label everything clearly, add directional indicators, include legend for symbols

5. Scale Mismatch

Symptoms: Wrong diagram type for audience, too technical for business users, too simplistic for developers
Detection Rule: If stakeholders ask "what does this mean" or "where's the detail"
Fix: Match diagram complexity to audience expertise, create multiple views for different stakeholders

Worked Examples

Example 1: Refactoring a Crowded System Architecture

Initial Problem: 15-microservice system crammed into one diagram

# BEFORE (Failure Mode: Information Overload)
[Cramped diagram with 15 services, 30+ arrows, unreadable labels]

# ANALYSIS: Too much complexity for single view
- Stakeholder: Engineering team (technical)
- Purpose: Understanding service dependencies  
- Current issues: Can't trace request flows, unclear boundaries

# DECISION PROCESS:
1. Problem size = Large (15 elements) → Multiple diagrams needed
2. Audience = Technical → Detailed views acceptable
3. Main concerns = Dependencies → Focus on interaction patterns

# SOLUTION: 3-tier approach

Overview Diagram:

┌─────────────────────────────────────────┐
│              API Gateway                │
└──────────────┬──────────────────────────┘
               │
┌──────────────┴──────────────────────────┐
│           Core Services (6)              │
│  ┌─────────┐ ┌─────────┐ ┌─────────┐   │
│  │ Auth    │ │ User    │ │ Order   │   │
│  └─────────┘ └─────────┘ └─────────┘   │
└──────────────┬──────────────────────────┘
               │
┌──────────────┴──────────────────────────┐
│          Support Services (9)            │
│     Logging, Metrics, Storage...         │
└─────────────────────────────────────────┘

Detail Diagram (Core Services):

    ┌─────────┐
    │   Auth  │◄─────────────┐
    └────┬────┘              │
         │ validates         │ requests
         ▼                   │ auth
    ┌─────────┐         ┌────┴────┐
    │  User   │◄────────┤ Order   │
    │ Service │ lookup  │ Service │
    └─────────┘         └─────────┘

Key Decisions Made:

  • Split by service tier (API → Core → Support)
  • Overview for navigation, details for implementation
  • Kept arrows minimal, focused on critical dependencies

Example 2: Choosing Between Diagram Types for Same System

Scenario: Need to show employee feedback process to different audiences

Decision Analysis:

Audience 1: HR Policy Team (non-technical)
→ Need: Understand policy compliance checkpoints
→ Choice: Process flowchart with decision diamonds

Audience 2: Engineering Team  
→ Need: Understand system integration points
→ Choice: Sequence diagram with API calls

Audience 3: Employees
→ Need: Understand what to expect
→ Choice: Simple linear timeline

HR Policy View (Process focus):

┌─────────────┐
│ Employee    │
│ Submits     │
│ Feedback    │
└──────┬──────┘
       │
       ▼
   ┌───────┐    No    ┌─────────────┐
   │Anonymous? ├──────►│ Route to    │
   │         │        │ Manager     │
   └───┬─────┘        └─────────────┘
       │ Yes
       ▼
┌─────────────┐
│ Route to    │
│ HR Only     │
└─────────────┘

Engineering View (System focus):

┌─────────┐      ┌─────────┐      ┌─────────┐
│ Web App │─────►│ API     │─────►│ Queue   │
└─────────┘ POST └─────────┘ pub  └─────────┘
                      │                │
                      │ auth           │ consume
                      ▼                ▼
               ┌─────────┐      ┌─────────┐
               │ Auth    │      │ Process │
               │ Service │      │ Worker  │
               └─────────┘      └─────────┘

Trade-off Analysis:

  • HR diagram prioritizes policy compliance over technical accuracy
  • Engineering diagram shows actual implementation flow
  • Both valid for their audiences - same system, different mental models

Quality Gates

Diagram Readability Test:

  • Can be understood by target audience in <30 seconds
  • No more than 7±2 elements per visual chunk
  • All text readable in monospace font at standard size
  • Clear visual hierarchy (primary/secondary/tertiary elements)

Content Coverage Checklist:

  • All significant elements labeled clearly
  • Directional flow indicated where applicable
  • Decision points explicitly marked
  • Relationships between elements are obvious

Layout Balance Verification:

  • Elements evenly distributed (no cramped corners)
  • Consistent spacing between similar elements
  • Arrow crossings minimized (<3 total)
  • Diagram fits in 80-character width when possible

Maintenance Acceptance Criteria:

  • Uses consistent character set throughout
  • Text labels can be updated without layout changes
  • Adding one element won't require complete redesign
  • Follows established pattern library conventions

Stakeholder Validation Gates:

  • Technical accuracy verified by domain expert
  • Complexity level appropriate for intended audience
  • Answers the specific question it was created for
  • Can stand alone without extensive verbal explanation

NOT-FOR Boundaries

Do NOT use this skill for:

  • Pixel-perfect UI mockups → Use design tools (Figma, Sketch)
  • Photo editing or image manipulation → Use image editing software
  • Complex mathematical visualizations → Use specialized math tools
  • Interactive animations → Use animation software
  • Publication-quality graphics → Use professional design tools

Delegation Guidelines:

  • For UI/UX wireframes → Use ux-designer skill instead
  • For data visualization with charts → Use data-analyst skill
  • For architectural blueprints → Use CAD software
  • For presentation graphics → Use presentation-designer skill
  • For infographic design → Use graphic design tools

Complexity Boundaries:

  • If diagram needs >50 elements → Break into multiple diagrams or use specialized tools
  • If requiring precise positioning → Use vector graphics software
  • If stakeholders need interactive exploration → Use digital diagramming tools
  • If output needs color for comprehension → Use tools that support color

Medium Limitations:

  • ASCII/Unicode text only - no images, colors, or fonts
  • Monospace character grid constraints
  • Static representation - no animation or interaction
  • Limited to characters available in standard Unicode sets

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Expert in 2000s-era music visualization (Milkdrop, AVS, Geiss) and modern WebGL implementations. Specializes in Butterchurn integration, Web Audio API AnalyserNode FFT data, GLSL shaders for audio-reactive visuals, and psychedelic generative art. Activate on "Milkdrop", "music visualization", "WebGL visualizer", "Butterchurn", "audio reactive", "FFT visualization", "spectrum analyzer". NOT for simple bar charts/waveforms (use basic canvas), video editing, or non-audio visuals.

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

Expert legal research agent for finding and scraping expungement data state by state. Knows authoritative sources, URL patterns, Firecrawl configuration, and 2026 legal landscape.

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

Expert in 3D computer vision labeling tools, workflows, and AI-assisted annotation for LiDAR, point clouds, and sensor fusion. Covers SAM4D/Point-SAM, human-in-the-loop architectures, and vertical-specific training strategies. Activate on '3D labeling', 'point cloud annotation', 'LiDAR labeling', 'SAM 3D', 'SAM4D', 'sensor fusion annotation', '3D bounding box', 'semantic segmentation point cloud'. NOT for 2D image labeling (use clip-aware-embeddings), general ML training (use ml-engineer), video annotation without 3D (use computer-vision-pipeline), or VLM prompt engineering (use prompt-engineer).

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

Apply crisis decision-making research to agent routing, uncertainty triage, and coordination failure analysis in time-pressured systems. Use when diagnosing handoff failures, analytical paralysis, or expert judgment under incomplete information. NOT for routine coding, simple CRUD design, or static single-agent tasks with complete information.

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

Use for insight, reframing, contradiction, impasse, and anomaly-driven problem solving when execution effort no longer helps. NOT for routine optimization, error correction, or well-specified tasks with known solution paths.

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

Apply cognitive task analysis to expert work that depends on perceptual cues, branching judgment, and recurring monitoring loops. Use when decomposing expert capability into agent structure, simulation design, or validation interviews. NOT for ordinary step-by-step SOP capture or simple pipelines with no tacit cue layer.

日本語の概要は準備中です。原文の説明を表示しています。

curiositech/port-daddy22026年10月12日 更新

curiositech のスキルをすべて見る

このスキルの問題を報告する