jerry-bot/claude-skills-reference/mermaid/SKILL.md

name, description, allowed-tools, metadata

name	description	allowed-tools	metadata
mermaid	Generate Mermaid diagrams from user requirements. Supports flowcharts, sequence diagrams, class diagrams, ER diagrams, Gantt charts, and 18 more diagram types.	Read Write Edit	argument-hint [diagram description or requirements]

Mermaid Diagram Generator

Generate high-quality Mermaid diagram code based on user requirements.

Workflow

1. Understand Requirements: Analyze user description to determine the most suitable diagram type
2. Read Documentation: Read the corresponding syntax reference for the diagram type
3. Generate Code: Generate Mermaid code following the specification
4. Apply Styling: Apply appropriate themes and style configurations

Diagram Type Reference

Select the appropriate diagram type and read the corresponding documentation:

Type	Documentation	Use Cases
Flowchart	flowchart.md	Processes, decisions, steps
Sequence Diagram	sequenceDiagram.md	Interactions, messaging, API calls
Class Diagram	classDiagram.md	Class structure, inheritance, associations
State Diagram	stateDiagram.md	State machines, state transitions
ER Diagram	entityRelationshipDiagram.md	Database design, entity relationships
Gantt Chart	gantt.md	Project planning, timelines
Pie Chart	pie.md	Proportions, distributions
Mindmap	mindmap.md	Hierarchical structures, knowledge graphs
Timeline	timeline.md	Historical events, milestones
Git Graph	gitgraph.md	Branches, merges, versions
Quadrant Chart	quadrantChart.md	Four-quadrant analysis
Requirement Diagram	requirementDiagram.md	Requirements traceability
C4 Diagram	c4.md	System architecture (C4 model)
Sankey Diagram	sankey.md	Flow, conversions
XY Chart	xyChart.md	Line charts, bar charts
Block Diagram	block.md	System components, modules
Packet Diagram	packet.md	Network protocols, data structures
Kanban	kanban.md	Task management, workflows
Architecture Diagram	architecture.md	System architecture
Radar Chart	radar.md	Multi-dimensional comparison
Treemap	treemap.md	Hierarchical data visualization
User Journey	userJourney.md	User experience flows
ZenUML	zenuml.md	Sequence diagrams (code style)

Configuration & Themes

• Theming - Custom colors and styles
• Directives - Diagram-level configuration
• Layouts - Layout direction and spacing
• Configuration - Global settings
• Math - LaTeX math support

Preferred Style (Flowcharts)

Arrow weight by meaning

Arrow	Use
==>	Main flow / happy path / player traffic
-->	API calls, decisions, branch outcomes
-.->	Async, background, retry loops

Color palette

Always include these classDefs in flowcharts:

classDef deploy   fill:#d5e8d4,stroke:#82b366,color:#000
classDef svc      fill:#dae8fc,stroke:#6c8ebf,color:#000
classDef pvc      fill:#f5f5f5,stroke:#666,color:#000
classDef external fill:#f5f5f5,stroke:#666,color:#000,stroke-dasharray:5 5
classDef ok       fill:#d5e8d4,stroke:#82b366,color:#000
classDef kick     fill:#f8cecc,stroke:#b85450,color:#000

Layout rules

• Natural depth — each terminal node gets its own exit point; no shared terminals across paths of different lengths. Dagre places nodes at max(predecessor depths) + 1.
• Side API calls — branch API call nodes off the calling node with --> at the same depth as the next decision node so they appear side-by-side, not inline in the ==> chain.
• Terminal naming — prefix ✓ for success, ✗ for failure (e.g. ✗ Kick: account locked).
• Section comments — use %% ── Section name ── to separate logical groups (traffic, auth, control plane, etc.).

Output Specification

Generated Mermaid code should:

1. Be wrapped in ```mermaid code blocks
2. Have correct syntax that renders directly
3. Have clear structure with proper line breaks and indentation
4. Use semantic node naming
5. Apply the preferred style above for flowcharts

Example Output

flowchart TD
    A[Start] --> B{Condition}
    B -->|Yes| C(["✓ Done"]):::ok
    B -->|No| D(["✗ Failed"]):::kick

    classDef ok   fill:#d5e8d4,stroke:#82b366,color:#000
    classDef kick fill:#f8cecc,stroke:#b85450,color:#000

---

User requirements: $ARGUMENTS