Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Generates Mermaid diagrams from 12 pre-built templates with LLM-powered placeholder filling. Supports architecture, flowchart, sequence, Gantt, class, and state diagram types with a consistent color scheme. A subagent handles the full workflow: reading source content, selecting templates, rendering diagrams, and validating output.
Rendering diagrams locally with mermaid-cli costs 95-99% less than AI image generation services while producing consistently styled, version-controllable vector output.
When to use it
- Generating system architecture diagrams for technical documentation
- Creating Gantt charts for project milestone tracking
- Producing UML class diagrams from code structure descriptions
- Building sequence diagrams for API interaction walkthroughs
- Visualizing process state machines in workflow documentation
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Generate 3 diagrams for chapters/chapter-05.md
- 1Reads chapter-05.md and analyzes its content
- 2Selects 3 appropriate templates from the 12 available types
- 3Fills each template's placeholders with labels derived from chapter content and writes content.json
- 4Runs generate.mjs to render .mmd source, .svg, and .png files
- 5Runs validate.mjs to confirm all outputs rendered without errors or unresolved placeholders
3 diagram sets saved to diagrams/chapter-05/ as .mmd, .svg, and .png files with a summary.json metadata file
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.