MCP HubMCP Hub
Вернуться к навыкам

mermaid-conventions

vamseeachanta
Обновлено Today
19 просмотров
3
2
3
Посмотреть на GitHub
Другое

О программе

Этот навык предоставляет соглашения для создания единообразных диаграмм Mermaid, уделяя особое внимание расположению узлов решений и порядку связей. Он обеспечивает визуальное выделение позитивного/успешного пути, размещая его слева за счёт перечисления первым в вертикальных блок-схемах. Используйте его для поддержания визуальной согласованности и ясности при создании блок-схем с точками принятия решений.

Быстрая установка

Claude Code

Рекомендуется
Основной
npx skills add vamseeachanta/workspace-hub
Команда плагинаАльтернативный
/plugin add https://github.com/vamseeachanta/workspace-hub
Git клонированиеАльтернативный
git clone https://github.com/vamseeachanta/workspace-hub.git ~/.claude/skills/mermaid-conventions

Скопируйте и вставьте эту команду в Claude Code для установки этого навыка

Документация

Mermaid Diagram Conventions

Decision Node Layout (Flowcharts)

When creating mermaid flowcharts with decision nodes ({}), follow these rules for edge ordering:

Rule 1: Yes/positive path goes LEFT (list first)

The first edge from a decision node renders to the left in top-down (TD) flowcharts. Always list the positive/affirmative/happy-path edge first.

%% CORRECT: Yes listed first → renders left (prominent)
flowchart TD
    A{Ready?}
    A -- Yes --> B[Proceed]
    A -- No --> C[Wait]
%% WRONG: No listed first → renders left (confusing)
flowchart TD
    A{Ready?}
    A -- No --> C[Wait]
    A -- Yes --> B[Proceed]

Rule 2: Happy path stays in the main vertical flow

The positive/expected path should continue straight down the main trunk of the diagram. Error handling, fallbacks, and edge cases should branch off to the side.

%% CORRECT: Happy path flows straight down
flowchart TD
    A[Start] --> B{Valid?}
    B -- Yes --> C[Process]
    B -- No --> D[Show error]
    C --> E[Done]

Rule 3: Negative/discouraging behavior appears later in hierarchy

Place blocking, error, and "don't do this" paths lower in the visual hierarchy:

  • Positive outcomes are rendered first (visually prominent, left side)
  • Negative outcomes are rendered second (visually subordinate, right side)
  • Dead-ends and error states appear at the leaves, not near the root

Rule 4: Edge ordering determines position

In mermaid, the order you list edges from a node controls layout:

  • First edge → left (TD) or top (LR)
  • Second edge → right (TD) or bottom (LR)

Use this to control which path gets visual prominence.

Subgraph Ordering

List subgraphs in logical reading order:

  1. Inputs/sources first (e.g., "Users", "Engineer Laptops")
  2. Processing/infrastructure middle (e.g., "Network", "Servers")
  3. Outputs/results last (e.g., "License Servers", "Results")

Connection Styles

Use connection styles to convey meaning:

  • --> solid arrow: active/primary connection
  • -.-> dotted arrow: blocked/inactive/secondary connection
  • -.- dotted line (no arrow): association/relationship (not a flow)
  • --- solid line (no arrow): strong association

PDF Rendering

When documents will be converted to PDF:

  • Mermaid blocks render via mermaid.ink API as PNG images
  • Keep diagrams simple enough to be readable at A4 scale
  • Use generate_pdfs.py in the admin directory to regenerate PDFs after mermaid changes
  • The script includes retry logic for mermaid.ink rate limits

Checklist

Before committing a mermaid diagram:

  • Yes/positive edges listed before No/negative edges at every decision node
  • Happy path flows straight down the main trunk
  • Error/blocking states branch to the side
  • Subgraphs in logical reading order
  • Connection styles match the relationship type
  • Diagram readable at A4 print size

GitHub репозиторий

vamseeachanta/workspace-hub
Путь: .claude/skills/_internal/documentation/mermaid-conventions

Похожие навыки

legal-sanity-review

Другое

This mandatory workflow scans code for legal risks like client names or proprietary references before cross-review. It blocks the review cycle if violations are found, requiring fixes first. Use it as a pre-gate for PR creation, code porting, or any cross-review trigger.

Просмотреть навык

brainstorming

Другое

The brainstorming skill transforms initial ideas into detailed specifications through iterative questioning and collaborative dialogue. It helps developers explore design options, evaluate trade-offs, and create shared understanding before implementation. This prevents rework by systematically refining concepts into actionable plans.

Просмотреть навык

repo-cleanup-progress-tracking-commands

Другое

This sub-skill provides shell commands for tracking file counts during repository cleanup operations, specifically for hidden folders like `.claude` and `.agent-os`. It includes verification scripts to ensure no files are lost after merging changes. Use it when consolidating or reorganizing project directories to monitor progress and validate results.

Просмотреть навык

brainstorming-error-handling

Другое

This Claude Skill provides structured guidance for handling common brainstorming roadblocks like conflicting requirements or scope creep. It offers specific action templates to resolve issues such as unclear responses or analysis paralysis. Developers should use this reference skill to maintain productive brainstorming sessions within Claude Code.

Просмотреть навык