drawio-academic-skills

Create publication-ready draw.io figures with academic policy gates and YAML-first workflows.

284|17|Updated Dec 26, 2025
One-click install
npx skills add https://github.com/bahayonghang/drawio-skills --skill drawio-academic-skills-bahayonghang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: drawio-academic-skills
Source: https://github.com/bahayonghang/drawio-skills/tree/main/skills/drawio-academic-skills
Command: npx skills add https://github.com/bahayonghang/drawio-skills --skill drawio-academic-skills-bahayonghang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Producing figures for papers, theses, IEEE submissions, and journals requires strict venue rules—print-safe palettes, caption and legend integrity, formula fidelity, and A4 readability—that generic diagram tools do not enforce. This overlay applies those academic policy gates on top of the sibling Draw.io Base Skill so publication figures pass venue constraints before export. ## Core Features & Use Cases - Academic preflight and policy gates: Determines venue, figure type (architecture, roadmap, workflow), palette with colorblind/grayscale safety, caption/legend needs, and print targets (cn-thesis, ieee-single, ieee-double) before rendering. - YAML-first offline workflow: Generates .drawio plus 300dpi PNG via draw.io Desktop (SVG fallback), with .spec.yaml and .arch.json sidecars kept in a work directory; no MCP or live backend required. - Publication playbooks and templates: Ships an academic figure playbook, export checklist, node-budget rules, paper examples (YOLO architecture, max-pooling, ablation pipeline, IEEE network), and compact templates. - Use Case: A researcher needs an IEEE single-column camera-ready architecture figure. The skill recommends the ieee-bw palette, validates print font floors, renders through the sibling base CLI, and exports a vector PDF via draw.io Desktop. ## Quick Start Ask the assistant to create an IEEE-style paper figure from your system description, for example: create a grayscale IEEE campus network architecture figure with core, distribution, and access layers and export it as .drawio and PDF.

Frequently Asked Questions about drawio-academic-skills

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

FAQPage Schema
How do I create an IEEE paper figure with draw.io?

Describe the figure and venue, and the skill runs an academic preflight that recommends a print-safe palette such as ieee-bw, sets meta.print.target to ieee-single or ieee-double, and renders through the sibling base CLI. For IEEE vector submission, export a PDF explicitly via draw.io Desktop since IEEE accepts PS/EPS/PDF only.

What is the difference between drawio base skill and the academic overlay?

The base skill owns shared execution: CLI, schemas, themes, palettes, and general workflows. The academic overlay owns publication policy, the figure playbook, export checklist, and paper examples, and depends on the sibling ../drawio directory for all rendering.

Does this skill work without draw.io Desktop installed?

Yes. The default offline deliverables are .drawio and SVG with sidecars in a work directory. PNG, PDF, and JPG exports require draw.io Desktop; without it the skill reports the missing export and can generate a diagrams.net browser URL as a fallback.

How are colorblind-safe and grayscale-safe palettes enforced?

After venue selection, the skill asks one palette question with the venue recommendation first, stating colorblind and grayscale safety per option. A grayscale-unsafe palette on a print target triggers PALETTE_PRINT_GATE, which strict validation escalates to an error recommending ieee-bw or tol-high-contrast.

Can I replicate an existing paper figure or screenshot into draw.io?

Yes. The replicate route rebuilds the figure with native draw.io cells, preserving captions, legends, formulas, callouts, and edge labels with explicit bounds and labelOffset. It sets meta.source to replicated and verifies the exported artifact before any browser screenshot.

What are the node count limits for academic figures?

The playbook targets under 40 nodes: 41-60 triggers a warning, 61-100 is blocked in strict mode, and over 100 is a hard limit. Compact templates and single-node legends help keep complex architectures within budget.