2.5k
GitHub Stars
6
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill visual-explainer-openclaw- _meta.json322 B
- CHANGELOG.md15.9 KB
- package.json266 B
- README.md2.8 KB
- skill.json952 B
- SKILL.md36.6 KB
Overview
This skill generates self-contained, browser-ready HTML pages that visually explain systems, code changes, plans, and data. It produces diagrams, styled tables, slide-like pages, and visual diff/plan reviews and always opens the result in a local browser. When complex tabular data or multi-section visualizations are needed, the output is an accessible, copy-paste-friendly HTML file saved to the workspace directory.
How this skill works
When given a request for a diagram, architecture overview, diff review, plan review, project recap, comparison table, or any visual explanation, the skill chooses an appropriate template and renders a fully styled HTML page. It prefers Mermaid for graphs and flowcharts, semantic HTML tables for data matrices, and CSS+Chart.js for dashboards, embedding assets inline so the page is self-contained. For long or wide tables it proactively generates the HTML page instead of ASCII art, writes it to ~/.openclaw/workspace/diagrams/, and opens it in the browser.
When to use it
- You need an architecture diagram, flowchart, sequence diagram, ER/schema, or state machine.
- A code diff or plan needs a visual comparison and risk assessment.
- You want a polished comparison table, feature matrix, or audit report (4+ rows or 3+ columns triggers HTML).
- Preparing a project recap or handoff page for context switching.
- Generating a slide-style visual brief or implementation plan.
- Converting complex ASCII tables into styled, copyable HTML pages
Best practices
- Pick the viewer and information density first (developer, PM, or exec) and let that drive layout complexity.
- Prefer constrained aesthetics (Blueprint, Editorial, Paper/ink) for consistent, readable output.
- Use Mermaid for topology and flow diagrams; use semantic <table> markup for data-heavy matrices.
- Embed fonts and assets via links and fallbacks; avoid Inter/Roboto/Helvetica as the primary body font.
- For diagrams with many nodes, either increase font/zoom or use a hybrid overview + detailed cards pattern.
Example use cases
- Visual diff review: side-by-side architecture diagrams with annotated risk notes.
- Plan review: implementation timeline with task cards, owners, and risk flags.
- Project recap: mental-model snapshot with component cards, recent changes, and next steps.
- Comparison table: feature matrix exported as a styled HTML table for stakeholder review.
- Slide deck: magazine-quality slide pages exported as single-file HTML for quick sharing.
FAQ
HTML pages are written to ~/.openclaw/workspace/diagrams/ and opened in your default local browser.
Will the skill ever return ASCII tables?
No. For complex tables or when this skill is active, the output is an HTML page; you may get a short chat summary alongside the file path.