ck:tech-graph

Generate SVG+PNG technical diagrams from natural language descriptions.

Updated Jun 14, 2026
One-click install
npx skills add https://github.com/duy-tung/claude-code-setup --skill ck-tech-graph
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:tech-graph
Source: https://github.com/duy-tung/claude-code-setup/tree/main/claude/skills/tech-graph
Command: npx skills add https://github.com/duy-tung/claude-code-setup --skill ck-tech-graph

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, pypdf, pdfplumber, librsvg, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill generates high-quality SVG+PNG technical diagrams from natural language descriptions, automating the creation of architecture, data flow, flowchart, sequence, agent/memory, or concept maps.

Core Features & Use Cases

  • Diagram Generation: Automatically generates diagrams based on natural language descriptions.
  • Multiple Styles: Supports 7 visual styles including Flat Icon, Dark Terminal, Blueprint, Notion Clean, Glassmorphism, Claude Official, and OpenAI Official.
  • Diagram Types: Offers support for 14 diagram types including architecture, data flow, flowchart, sequence, comparison, timeline, mind map, and UML diagrams.
  • AI/Agent Domain Patterns: Built-in support for AI/agent domain patterns like RAG, Agentic Search, Mem0, Multi-Agent, and Tool Call.
  • Semantic Shape Vocabulary: Consistent use of shapes to represent different concepts.
  • SVG + PNG Output: Generates SVG files for editing and PNG files for embedding.
  • AI/ML Product Coverage: Covers popular AI/ML products and frameworks.

Quick Start

Use the ck:tech-graph skill to generate a diagram. For example: "Generate a microservices architecture diagram in style 2 (Dark Terminal)".

Frequently Asked Questions about ck:tech-graph

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate technical diagrams from natural language descriptions?

To generate technical diagrams from natural language, you describe the architecture or data flow in plain text. The system processes this input to output production-quality SVG and PNG files, supporting 14 diagram types including UML, sequence, and flowcharts.

Can I use Python to automatically create UML diagrams for AI agent systems?

Yes, you can use Python to automatically create UML diagrams for AI agent systems. The tool requires Python along with pypdf, pdfplumber, and librsvg, providing built-in patterns for RAG, Multi-Agent, and Tool Call visualizations.

What visual styles are available for automated system visualization?

Automated system visualization supports 7 visual styles including Dark Terminal, Blueprint, Glassmorphism, Notion Clean, Flat Icon, Claude Official, and OpenAI Official. These styles apply consistent semantic shape vocabularies across various diagram types.

Does generating SVG and PNG architecture diagrams require external rendering libraries?

Yes, generating SVG and PNG architecture diagrams requires external rendering libraries. The environment must include Python, pypdf, pdfplumber, and librsvg to process the semantic shape vocabulary and render the final visual outputs.

What is the best way to document microservices architecture using AI?

The best way to document microservices architecture using AI is generating editable SVG and embeddable PNG files directly from text descriptions. This approach automates architectural design visualization without manual diagramming tools.

When should I use natural language diagram generation instead of manual drawing tools?

You should use natural language diagram generation instead of manual drawing tools when creating production-quality technical documentation for AI/ML products. It automates complex system visualization, ensuring consistent semantic shapes across 14 diagram types.