Design Doc Mermaid logo

Design Doc Mermaid

Organization
SpillwaveSolutions
design-doc-mermaid

Create Mermaid diagrams (flowchart, sequence, class, ER, state, C4, architecture) from text or source code. Default for GitHub wiki. Use when asked to create a diagram, generate mermaid, document architecture, or convert code to diagram. PlantUML is only for leftover types. Confluence needs PNG/SVG as well as the fence.

Overview

PublisherSpillwaveSolutions
Repositorydesign-doc-mermaid
Skill namedesign-doc-mermaid
Stars
176
Forks
25
Bundled files
18
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 18 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by SpillwaveSolutions on GitHub. Read the source before you install it.

Installation

Install the Design Doc Mermaid AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/SpillwaveSolutions/design-doc-mermaid/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SpillwaveSolutions/design-doc-mermaid.git \
  .claude/skills/design-doc-mermaid
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Design Doc Mermaid in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Design Doc Mermaid on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Design Doc Mermaid is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Mermaid Architect - Hierarchical Diagram and Documentation Skill

Mermaid diagram and documentation system with specialized guides and code-to-diagram capabilities.

Table of Contents

Decision Tree

How this skill works:

  1. User makes a request → Skill analyzes intent
  2. Skill determines diagram/document type → Loads appropriate guide(s)
  3. AI reads specialized guide → Generates diagram/document using templates
  4. Result delivered → With validation and export options

User Intent Analysis:

Available Guides and Resources

Diagram Type Guides (references/guides/diagrams/)

GuideFull PathLoad When User WantsExamples
Activity Diagramsreferences/guides/diagrams/activity-diagrams.mdWorkflows, processes, business logic, user flows, decision trees"Show checkout flow", "Document ETL pipeline", "Create approval workflow"
Deployment Diagramsreferences/guides/diagrams/deployment-diagrams.mdInfrastructure, cloud architecture, K8s, serverless, network topology"Show AWS architecture", "Document GCP deployment", "Create K8s diagram"
Architecture Diagramsreferences/guides/diagrams/architecture-diagrams.mdSystem architecture, component design, high-level structure"Show system components", "Document microservices", "Architecture overview"
Sequence Diagramsreferences/guides/diagrams/sequence-diagrams.mdAPI interactions, service communication, request/response flows"Show API call sequence", "Document auth flow", "Service interactions"
WikiTicket / GitHub / Confluencereferences/guides/wiki-ticket-and-github.mdDesign docs, walkthroughs, requirements, wiki publish, Confluence images"architecture doc", "code walkthrough", "wiki", "Confluence"

Default for WikiTicket and GitHub wiki is this skill, including classDiagram, erDiagram, and stateDiagram-v2. Do not send those to PlantUML. If GitHub fails to render C4 / architecture-beta / block-beta, fall back to flowchart TD.

Code-to-Diagram Guide & Examples

ResourceFull PathWhat It Provides
Master Guidereferences/guides/code-to-diagram/README.mdComplete workflow for analyzing any codebase and extracting diagrams
Spring Bootexamples/spring-boot/README.mdController→Service→Repository architecture, deployment config, sequence from methods, activity from business logic
FastAPIexamples/fastapi/README.mdPython async patterns, Pydantic models, dependency injection, cloud deployment
Reactexamples/react/README.mdComponent hierarchy, state management, data flow, build pipeline
Python ETLexamples/python-etl/README.mdData pipeline, transformation steps, error handling, scheduling
Node/Expressexamples/node-webapp/README.mdMiddleware chain, route handlers, async patterns, deployment
Java Web Appexamples/java-webapp/README.mdTraditional MVC, servlet containers, WAR deployment

Design Document Templates

TemplateFull PathUse ForLoad When
Architecture Designassets/architecture-design-template.mdSystem-wide architecture"Create architecture doc", "Document system design"
API Designassets/api-design-template.mdAPI specifications"API design doc", "Document REST API"
Feature Designassets/feature-design-template.mdFeature planning"Feature design", "Plan new feature"
Database Designassets/database-design-template.mdDatabase schema"Database design", "Document schema"
System Designassets/system-design-template.mdComplete system"System design doc", "Full system documentation"

Unicode Symbols Guide

Full Path: references/guides/unicode-symbols/guide.md

Load when user mentions: "unicode symbols", "emoji in diagrams", "semantic icons", "add symbols"

Quick Reference:

  • 📦 Infrastructure: ☁️ 🌐 🔌 📡 🗄️
  • ⚙️ Compute: ⚙️ ⚡ 🔄 ♻️ 🚀 💨
  • 💾 Data: 💾 📦 📊 📈 🗃️ 🧊
  • 📨 Messaging: 📨 📬 📤 📥 🐰 📢
  • 🔐 Security: 🔐 🔑 🛡️ 🚪 👤 🎫
  • 📝 Monitoring: 📝 📊 🚨 ⚠️ ✅ ❌

Python Scripts (scripts/)

ScriptUse ForLoad When
extract_mermaid.pyExtract diagrams from Markdown, validate syntax, replace with images"extract diagrams", "validate mermaid", "find all diagrams"
mermaid_to_image.pyConvert .mmd to PNG/SVG, batch conversion, custom themes"convert to image", "render diagram", "create PNG"
resilient_diagram.pyFull workflow: save .mmd, generate image, validate, error recovery"generate diagram", "create diagram with validation", "resilient diagram"

GitHub wiki, WikiTicket, and Confluence

Load references/guides/wiki-ticket-and-github.md for WikiTicket design docs, code walkthroughs, requirements, GitHub wiki, or Confluence.

TargetWhat to ship
GitHub wiki / GFMFenced mermaid block. GitHub renders class, ER, state, sequence, flowchart, C4.
Confluence / Notion / Word / PDFAlso render PNG or SVG (scripts/resilient_diagram.py or mmdc) and upload the image. Do not rely on Confluence to render Mermaid.

PlantUML is opt-in for Salt wireframes, use case, timing, ArchiMate, nwdiag, and WBS. Those always ship as PNG or SVG.

Usage Patterns

Common request patterns and guide selection. See When to Use What for complete mapping.

PatternExample RequestGuides to Load
Single Diagram"Create activity diagram for login flow"Diagram type guide + Unicode symbols
Code-to-Diagram"Generate deployment from application.yml"Framework example + Deployment guide
Design Document"Create API design document"Template from assets/ + Relevant diagram guides
Extract/Validate"Extract diagrams from design.md"Use scripts/extract_mermaid.py
Batch Convert"Convert all .mmd to PNG"Use scripts/mermaid_to_image.py

Resilient Workflow

CRITICAL: This is the recommended approach for ALL diagram generation. It ensures validation, error recovery, and consistent file organization.

Full Guide: references/guides/resilient-workflow.md

Workflow Overview

Key Principle

NEVER add a diagram to markdown until it passes validation. This prevents broken diagrams in documentation.

Using the Script (Recommended)

bash
# Generate with full error recovery
python scripts/resilient_diagram.py \
    --code "flowchart TD; A-->B" \
    --markdown-file design_doc \
    --diagram-num 1 \
    --title "process_flow" \
    --format png \
    --json

Output: Both .mmd and .png files in ./diagrams/ directory.

File Naming Convention

./diagrams/<markdown_file>_<num>_<type>_<title>.mmd
./diagrams/<markdown_file>_<num>_<type>_<title>.png

Example: ./diagrams/api_design_01_sequence_auth_flow.png

Error Recovery Priority

When validation fails, the workflow automatically:

  1. Check troubleshooting guide - references/guides/troubleshooting.md (28 documented errors)
  2. Search with perplexity - perplexity_ask MCP for syntax questions
  3. Search with brave - brave_web_search MCP for recent solutions
  4. Ask gemini - gemini skill for alternative perspective
  5. General search - WebSearch tool as fallback

Manual Fallback Steps

If the script is unavailable:

  1. Identify diagram type from first line (flowchart, sequence, etc.)
  2. Load reference guide from references/guides/diagrams/
  3. Save to ./diagrams/<markdown_file>_<num>_<type>_<title>.mmd
  4. Validate: mmdc -i file.mmd -o file.png -b transparent
  5. On error: Search references/guides/troubleshooting.md for matching error
  6. If not found: Use search tools in priority order above
  7. Add reference: ![Description](./diagrams/filename.png)

Pattern 6: Resilient Diagram Generation

User: "Create a sequence diagram and add it to the design doc"

Skill Actions:

  1. Identify intent: diagram generation + markdown integration
  2. Load workflow guide: references/guides/resilient-workflow.md
  3. Identify diagram type: sequence
  4. Load diagram guide: references/guides/diagrams/sequence-diagrams.md
  5. Generate Mermaid code using templates
  6. Execute resilient workflow:
    bash
    python scripts/resilient_diagram.py \
        --code "[generated code]" \
        --markdown-file design_doc \
        --diagram-num 1 \
        --title "api_sequence" \
        --json
  7. If validation fails → Apply troubleshooting fix → Retry
  8. On success → Add ![API Sequence](./diagrams/design_doc_01_sequence_api_sequence.png) to markdown

Unicode Semantic Symbols

Always use Unicode symbols to enhance diagram clarity. Common patterns:

Infrastructure & Deployment

Activity Flow with States

Microservices Architecture

For complete symbol reference, load: references/guides/unicode-symbols/guide.md

Python Utilities

Extract Mermaid Diagrams

bash
# List all diagrams
python scripts/extract_mermaid.py document.md --list-only

# Extract to separate files
python scripts/extract_mermaid.py document.md --output-dir diagrams/

# Validate all diagrams
python scripts/extract_mermaid.py document.md --validate

# Replace with image references (for Confluence upload)
python scripts/extract_mermaid.py document.md --replace-with-images \
  --image-format png --output-markdown output.md

Convert to Images

bash
# Single conversion
python scripts/mermaid_to_image.py diagram.mmd output.png

# With custom settings
python scripts/mermaid_to_image.py diagram.mmd output.svg \
  --theme dark --background white --width 1200

# Batch convert directory
python scripts/mermaid_to_image.py diagrams/ output/ --format png --recursive

# From stdin
echo "graph TD; A-->B" | python scripts/mermaid_to_image.py - output.png

Decision Tree Examples

Example 1: User Asks for Workflow Diagram

Input: "Show the checkout process workflow"

Skill Decision Path:

1. Analyze: workflow, process → ACTIVITY DIAGRAM
2. Load guide: guides/diagrams/activity-diagrams.md
3. Find pattern: E-commerce checkout (template exists in guide)
4. Generate using template + Unicode symbols
5. Output activity diagram with decision points

Output: Complete activity diagram with Unicode symbols for cart, payment, order states.

Example 2: User Provides Spring Boot Code

Input: "Here's my Spring Boot controller, create diagrams"

Skill Decision Path:

1. Analyze: Spring Boot, code provided → CODE-TO-DIAGRAM + SPRING BOOT
2. Load guides:
   - examples/spring-boot/README.md
   - guides/diagrams/architecture-diagrams.md (for structure)
   - guides/diagrams/sequence-diagrams.md (for method calls)
   - guides/diagrams/activity-diagrams.md (for business logic)
3. Generate multiple diagrams:
   a. Architecture diagram from @RestController/@Service/@Repository annotations
   b. Sequence diagram from method call chain
   c. Activity diagram from business logic flow
4. Output all diagrams with explanations

Output: 3-4 diagrams showing different views of the Spring Boot application.

Example 3: User Wants Infrastructure Documentation

Input: "Document my GCP Cloud Run deployment with AlloyDB"

Skill Decision Path:

1. Analyze: infrastructure, GCP, Cloud Run → DEPLOYMENT DIAGRAM
2. Load guides:
   - guides/diagrams/deployment-diagrams.md
   - examples/spring-boot/ or examples/fastapi/ (if code provided)
3. Check for IaC files (Pulumi, Terraform, docker-compose)
4. Generate deployment diagram with:
   - Cloud Run services with specs
   - VPC connector
   - AlloyDB cluster
   - Security (IAM, Secret Manager)
   - Monitoring
5. Apply Unicode symbols for clarity
6. Output with resource specifications

Output: Complete GCP deployment diagram with all resources labeled.

High-Contrast Styling

ALL diagrams MUST use high-contrast colors:

Rules:

  • Light background → Dark text color
  • Dark background → Light text color
  • Always specify color: in every classDef

File Organization

design-doc-mermaid/
├── SKILL.md                          # This file - Main orchestrator
├── README.md                         # User documentation
├── CLAUDE.md                         # Claude Code instructions
│
├── references/                       # Reference materials
│   ├── mermaid-diagram-guide.md     # Legacy general guide
│   └── guides/                       # Specialized guides (load on-demand)
│       ├── diagrams/
│       │   ├── activity-diagrams.md      # Workflows, processes
│       │   ├── deployment-diagrams.md    # Infrastructure, cloud
│       │   ├── architecture-diagrams.md  # System architecture
│       │   └── sequence-diagrams.md      # API interactions
│       ├── code-to-diagram/
│       │   └── README.md                 # Master guide for code analysis
│       ├── unicode-symbols/
│       │   └── guide.md                  # Complete symbol reference
│       └── troubleshooting.md        # Common syntax errors & fixes
│
├── assets/                           # Design document templates
│   ├── architecture-design-template.md
│   ├── api-design-template.md
│   ├── feature-design-template.md
│   ├── database-design-template.md
│   └── system-design-template.md
│
├── scripts/                          # Python utilities
│   ├── extract_mermaid.py           # Extract & validate diagrams
│   └── mermaid_to_image.py          # Convert to PNG/SVG
│
├── examples/                         # Language-specific patterns
│   ├── spring-boot/                 # Spring Boot patterns
│   ├── fastapi/                     # FastAPI patterns
│   ├── react/                       # React patterns
│   ├── python-etl/                  # Data pipeline patterns
│   ├── node-webapp/                 # Express.js patterns
│   └── java-webapp/                 # Traditional Java patterns
│
└── references/                       # General Mermaid reference
    └── mermaid-diagram-guide.md     # Complete Mermaid syntax guide

Workflow Summary

  1. Analyze user intent → Determine diagram type, document type, or action needed
  2. Load appropriate guide(s) → Read only what's needed (token efficient)
  3. Apply templates and patterns → Use examples from guides
  4. Generate output → Create diagram or document
  5. Validate (optional) → Use scripts to verify
  6. Convert (optional) → Export to images if needed

When to Use What

User RequestLoad This
"activity diagram", "workflow", "process flow"references/guides/diagrams/activity-diagrams.md
"deployment", "infrastructure", "cloud", "k8s"references/guides/diagrams/deployment-diagrams.md
"architecture", "system design", "components"references/guides/diagrams/architecture-diagrams.md + design template
"API", "sequence", "interactions", "flow"references/guides/diagrams/sequence-diagrams.md
"class diagram", "ER", "state machine", "wiki", "walkthrough", "architecture doc"references/guides/wiki-ticket-and-github.md plus the matching type guide
"Confluence", "upload image", "PNG", "SVG"scripts/resilient_diagram.py or scripts/mermaid_to_image.py
"Spring Boot code"examples/spring-boot/ + relevant diagram guides
"FastAPI code", "Python API"examples/fastapi/ + relevant diagram guides
"React app", "frontend"examples/react/ + architecture guide
"ETL", "data pipeline", "Python batch"examples/python-etl/ + activity guide
"symbols", "unicode", "emoji"references/guides/unicode-symbols/guide.md
"syntax error", "diagram won't render", "troubleshoot"references/guides/troubleshooting.md
"extract diagrams"scripts/extract_mermaid.py
"convert to image", "PNG", "SVG"scripts/mermaid_to_image.py
"create diagram", "generate diagram", "add diagram to markdown"scripts/resilient_diagram.py + references/guides/resilient-workflow.md
"design document", "full docs"assets/*-design-template.md + diagram guides

Best Practices

  1. Single Responsibility: One diagram = One concept
  2. Unicode Enhancement: Always use semantic symbols for clarity
  3. High Contrast: Never skip the color: property in styles
  4. Validate Early: Use scripts to catch syntax errors
  5. Template Reuse: Leverage existing templates and examples
  6. Load On-Demand: Only read guides needed for the specific request
  7. Token Efficiency: Use hierarchical loading instead of reading everything

Learning Path

New to Mermaid? Start here:

  1. Read references/guides/unicode-symbols/guide.md for symbol meanings
  2. Read references/guides/diagrams/activity-diagrams.md for basic patterns
  3. Try examples in examples/spring-boot/ or examples/fastapi/
  4. Use scripts/extract_mermaid.py --validate to check your work

Need to document code? Follow this:

  1. Identify your framework → Load relevant examples/{framework}/
  2. Match code pattern to diagram type
  3. Use templates from guide
  4. Validate with scripts

Creating design docs? Follow this:

  1. Choose document type → Load template from assets/
  2. Fill in text sections
  3. Load diagram guides as needed for each section
  4. Use Unicode symbols throughout
  5. Save to docs/design/ with timestamp

Version: 2.0 (Hierarchical Architecture) Last Updated: 2025-01-13 Maintained by: Claude Code Skills

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Design Doc Mermaid AI skill do?

Create Mermaid diagrams (flowchart, sequence, class, ER, state, C4, architecture) from text or source code. Default for GitHub wiki. Use when asked to create a diagram, generate mermaid, document architecture, or convert code to diagram. PlantUML is only for leftover types. Confluence needs PNG/SVG as well as the fence.

Why use Design Doc Mermaid on TypingMind?

Because you install it once and use it with any model. Design Doc Mermaid is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Design Doc Mermaid in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SpillwaveSolutions/design-doc-mermaid/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Design Doc Mermaid?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Design Doc Mermaid?

As many as you like. As long as a model supports skills, you can use Design Doc Mermaid with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Design Doc Mermaid AI skill free?

It is published on GitHub by SpillwaveSolutions. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇