本文へ移動
cccskills
無料GitHub で公開

mermaid-visualizer

Transform text content into professional Mermaid diagrams for presentations and documentation. Use when users ask to visualize concepts, create flowcharts, or make diagrams from text. Supports process flows, system architectures, comparisons, mindmaps, and more with built-in syntax error prevention.

インストール方法を見る

含まれるファイル(2)

  • SKILL.md8.7 KB
  • references/syntax-rules.md9.8 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。

Mermaid Visualizer

Overview

Convert text content into clean, professional Mermaid diagrams optimized for presentations and documentation. Automatically handles common syntax pitfalls (list syntax conflicts, subgraph naming, spacing issues) to ensure diagrams render correctly in Obsidian, GitHub, and other Mermaid-compatible platforms.

Quick Start

When creating a Mermaid diagram:

  1. Analyze the content - Identify key concepts, relationships, and flow
  2. Choose diagram type - Select the most appropriate visualization (see Diagram Types below)
  3. Select configuration - Determine layout, detail level, and styling
  4. Generate diagram - Create syntactically correct Mermaid code
  5. Output in markdown - Wrap in proper code fence with optional explanation

Default assumptions:

  • Vertical layout (TB) unless horizontal requested
  • Medium detail level (balanced between simplicity and information)
  • Professional color scheme with semantic colors
  • Obsidian/GitHub compatible syntax

Diagram Types

1. Process Flow (graph TB/LR)

Best for: Workflows, decision trees, sequential processes, AI agent architectures

Use when: Content describes steps, stages, or a sequence of actions

Key features:

  • Swimlanes via subgraph for grouping related steps
  • Arrow labels for transitions
  • Feedback loops and branches
  • Color-coded stages

Configuration options:

  • layout: "vertical" (TB), "horizontal" (LR)
  • detail: "simple" (core steps only), "standard" (with descriptions), "detailed" (with annotations)
  • style: "minimal", "professional", "colorful"

2. Circular Flow (graph TD with circular layout)

Best for: Cyclic processes, continuous improvement loops, agent feedback systems

Use when: Content emphasizes iteration, feedback, or circular relationships

Key features:

  • Central hub with radiating elements
  • Curved feedback arrows
  • Clear cycle indicators

3. Comparison Diagram (graph TB with parallel paths)

Best for: Before/after comparisons, A vs B analysis, traditional vs modern systems

Use when: Content contrasts two or more approaches or systems

Key features:

  • Side-by-side layout
  • Central comparison node
  • Clear differentiation via color/style

4. Mindmap

Best for: Hierarchical concepts, knowledge organization, topic breakdowns

Use when: Content is hierarchical with clear parent-child relationships

Key features:

  • Radial tree structure
  • Multiple levels of nesting
  • Clean visual hierarchy

5. Sequence Diagram

Best for: Interactions between components, API calls, message flows

Use when: Content involves communication between actors/systems over time

Key features:

  • Timeline-based layout
  • Clear actor separation
  • Activation boxes for processes

6. State Diagram

Best for: System states, status transitions, lifecycle stages

Use when: Content describes states and transitions between them

Key features:

  • Clear state nodes
  • Labeled transitions
  • Start and end states

Critical Syntax Rules

Always follow these rules to prevent parsing errors:

Rule 1: Avoid List Syntax Conflicts

❌ WRONG: [1. Perception]       → Triggers "Unsupported markdown: list"
✅ RIGHT: [1.Perception]         → Remove space after period
✅ RIGHT: [① Perception]         → Use circled numbers (①②③④⑤⑥⑦⑧⑨⑩)
✅ RIGHT: [(1) Perception]       → Use parentheses
✅ RIGHT: [Step 1: Perception]   → Use "Step" prefix

Rule 2: Subgraph Naming

❌ WRONG: subgraph AI Agent Core  → Space in name without quotes
✅ RIGHT: subgraph agent["AI Agent Core"]  → Use ID with display name
✅ RIGHT: subgraph agent          → Use simple ID only

Rule 3: Node References

❌ WRONG: Title --> AI Agent Core  → Reference display name directly
✅ RIGHT: Title --> agent          → Reference subgraph ID

Rule 4: Special Characters in Node Text

✅ Use quotes for text with spaces: ["Text with spaces"]
✅ Escape or avoid: quotation marks → use 『』 instead
✅ Escape or avoid: parentheses → use 「」 instead
✅ Line breaks in circle nodes only: ((Text<br/>Break))

Rule 5: Arrow Types

  • --> solid arrow
  • -.-> dashed arrow (for supporting systems, optional paths)
  • ==> thick arrow (for emphasis)
  • ~~~ invisible link (for layout only)

For complete syntax reference and edge cases, see references/syntax-rules.md

Configuration Options

All diagrams accept these parameters:

Layout:

  • direction: "vertical" (TB), "horizontal" (LR), "right-to-left" (RL), "bottom-to-top" (BT)
  • aspect: "portrait" (default), "landscape" (wide), "square"

Detail Level:

  • simple: Core elements only, minimal labels
  • standard: Balanced detail with key descriptions (default)
  • detailed: Full annotations, explanations, and metadata
  • presentation: Optimized for slides (larger text, fewer details)

Style:

  • minimal: Monochrome, clean lines
  • professional: Semantic colors, clear hierarchy (default)
  • colorful: Vibrant colors, high contrast
  • academic: Formal styling for papers/documentation

Additional Options:

  • show_legend: true/false - Include color/symbol legend
  • numbered: true/false - Add sequence numbers to steps
  • title: string - Add diagram title

Example Usage Patterns

Pattern 1: Basic request

User: "Visualize the software development lifecycle"
Response: [Analyze → Choose graph TB → Generate with standard detail]

Pattern 2: With configuration

User: "Create a horizontal flowchart of our sales process with lots of detail"
Response: [Analyze → Choose graph LR → Generate with detailed level]

Pattern 3: Comparison

User: "Compare traditional AI vs AI agents"
Response: [Analyze → Choose comparison layout → Generate with contrasting styles]

Workflow

  1. Understand the content

    • Identify main concepts, entities, and relationships
    • Determine hierarchy or sequence
    • Note any comparisons or contrasts
  2. Select diagram type

    • Match content structure to diagram type
    • Consider user's presentation context
    • Default to process flow if ambiguous
  3. Choose configuration

    • Apply user-specified options
    • Use sensible defaults for unspecified options
    • Optimize for readability
  4. Generate Mermaid code

    • Follow all syntax rules strictly
    • Use semantic naming (descriptive IDs)
    • Apply consistent styling
    • Test for common errors:
      • No "number. space" patterns in node text
      • All subgraphs use ID["display name"] format
      • All node references use IDs not display names
  5. Output with context

    • Wrap in ```mermaid code fence
    • Add brief explanation of diagram structure
    • Mention rendering compatibility (Obsidian, GitHub, etc.)
    • Offer to adjust or create variations

Color Scheme Defaults

Standard professional palette:

  • Green (#d3f9d8/#2f9e44): Input, perception, start states
  • Red (#ffe3e3/#c92a2a): Planning, decision points
  • Purple (#e5dbff/#5f3dc4): Processing, reasoning
  • Orange (#ffe8cc/#d9480f): Actions, tool usage
  • Cyan (#c5f6fa/#0c8599): Output, execution, results
  • Yellow (#fff4e6/#e67700): Storage, memory, data
  • Pink (#f3d9fa/#862e9c): Learning, optimization
  • Blue (#e7f5ff/#1971c2): Metadata, definitions, titles
  • Gray (#f8f9fa/#868e96): Neutral elements, traditional systems

Common Patterns

Swimlane Pattern (Grouping)

graph TB
    subgraph core["Core Process"]
        A --> B --> C
    end
    subgraph support["Supporting Systems"]
        D
        E
    end
    core -.-> support

Feedback Loop Pattern

graph TB
    A[Start] --> B[Process]
    B --> C[End]
    C -.->|Feedback| A

Hub and Spoke Pattern

graph TB
    Central[Hub]
    A[Spoke 1] --> Central
    B[Spoke 2] --> Central
    C[Spoke 3] --> Central

Quality Checklist

Before outputting, verify:

  • No "number. space" patterns in any node text
  • All subgraphs use proper ID syntax
  • All arrows use correct syntax (-->, -.->)
  • Colors applied consistently
  • Layout direction specified
  • Style declarations present
  • No ambiguous node references
  • Compatible with Obsidian/GitHub renderers
  • No Emoji in any node text - use text labels or color coding instead

References

For detailed syntax rules and troubleshooting, see:

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

defuddle

無料

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page. Do NOT use for URLs ending in .md — those are already markdown, use WebFetch directly.

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

Generate Excalidraw diagrams from text content. Supports three output modes - Obsidian (.md), Standard (.excalidraw), and Animated (.excalidraw with animation order). Triggers on "Excalidraw", "diagram", "standard excalidraw", "animate".

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

Create and edit JSON Canvas files (.canvas) with nodes, edges, groups, and connections. Use when working with .canvas files, creating visual canvases, mind maps, flowcharts, or when the user mentions Canvas files in Obsidian.

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

Interact with Obsidian vaults using the Obsidian CLI to read, create, search, and manage notes, tasks, properties, and more. Also supports plugin and theme development with commands to reload plugins, run JavaScript, capture errors, take screenshots, and inspect the DOM. Use when the user asks to interact with their Obsidian vault, manage notes, search vault content, perform vault operations from the command line, or develop and debug Obsidian plugins and themes.

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian-specific syntax. Use when working with .md files in Obsidian, or when the user mentions wikilinks, callouts, frontmatter, tags, embeds, or Obsidian notes.

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。

breferrari/obsidian-mind5,2802026年10月13日 更新

breferrari のスキルをすべて見る

このスキルの問題を報告する