classdiagram-image-to-json

Convert UML class diagram images into a machine-readable domain-model.json.

Updated Feb 19, 2026
One-click install
npx skills add https://github.com/atanaka/uml-workflow-v3 --skill classdiagram-image-to-json
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: classdiagram-image-to-json
Source: https://github.com/atanaka/uml-workflow-v3/tree/main/skills/classdiagram-image-to-json
Command: npx skills add https://github.com/atanaka/uml-workflow-v3 --skill classdiagram-image-to-json

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Convert hand-drawn or modeling tool UML class diagrams from images (JPEG, PNG, PDF) into a structured domain-model.json format to feed into the uml-workflow-v2 pipeline. It also supports multi-language OCR (Japanese/English) and merging with an existing model to keep workflows consistent.

Core Features & Use Cases

  • Multi-format extraction: supports JPEG, PNG, and PDF inputs to extract classes, attributes, methods, and relationships.
  • Merge capability: creates or updates domain-model.json and preserves metadata to maintain traceability.
  • Validation & integration: outputs a standards-compliant domain-model.json and integrates seamlessly with Step 6+ of the UML workflow.
  • Multi-language OCR: detects Japanese and English text to generate appropriate entity names and japanese_name fields.

Quick Start

Upload a class diagram image or PDF and request conversion to domain-model.json.

Frequently Asked Questions about classdiagram-image-to-json

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

FAQPage Schema
How do I convert a UML class diagram image into a JSON file?▼

To convert a UML class diagram image into a JSON file, upload your JPEG, PNG, or PDF diagram. The tool uses OCR to extract classes, attributes, methods, and relationships, outputting a structured domain-model.json.

Can I extract domain models from Draw.io or Lucidchart class diagrams?▼

Yes, you can extract domain models from Draw.io or Lucidchart class diagrams. The tool processes images exported from these modeling tools, reading UML structures to generate a valid domain-model.json format.

Does the class diagram OCR support Japanese text extraction?▼

The class diagram OCR supports Japanese text extraction. It detects both Japanese and English text within the diagram, generating appropriate entity names and populating the japanese_name fields in the output JSON.

How do I merge an extracted class diagram into an existing domain model?▼

You can merge an extracted class diagram into an existing domain model by providing the current domain-model.json. The conversion process updates the model while preserving metadata to maintain workflow traceability.

What is the best way to convert hand-drawn UML diagrams for software engineering workflows?▼

The best way to convert hand-drawn UML diagrams for software engineering workflows is to scan or upload the drawing as an image or PDF. The tool extracts the UML structures and outputs a standards-compliant domain-model.json for pipeline integration.

Do I need a specific OCR tool to process UML class diagrams with Japanese and English text?▼

You do not need a separate OCR tool to process UML class diagrams with Japanese and English text. The conversion process includes multi-language OCR capabilities that automatically detect and extract both languages into the JSON output.