skills.rest
Lingjie-wang
Lingjie-wangCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

render-html

Convert ARIS Markdown and JSON artifacts into single-file HTML views with templates.

Updated Jul 9, 2026
Search Tags:#template#markdown#html#single-file#report#render#research artifacts
One-click install
npx skills add https://github.com/Lingjie-wang/autoRL --skill render-html-lingjie-wang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: render-html
Source: https://github.com/Lingjie-wang/autoRL/tree/main/Auto-claude-code-research-in-sleep/skills/render-html
Command: npx skills add https://github.com/Lingjie-wang/autoRL --skill render-html-lingjie-wang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? ARIS research workflows produce Markdown artifacts (idea reports, auto-reviews, kill arguments, paper plans) that are machine-friendly but hard for humans to read. This Skill renders selected artifacts into polished single-file HTML pages with sticky tables of contents, callouts, math, and syntax highlighting, while keeping the Markdown as the canonical source. ## Core Features & Use Cases - Single-file HTML rendering: A pure-stdlib Python helper converts Markdown/JSON into one self-contained HTML file using academic or dashboard templates, with MathJax and highlight.js loaded from CDN (or skipped via --offline). - Drift detection and safety: Every output embeds the source path, SHA256, and timestamp in meta tags, and a sanitizer strips scripts, event handlers, and unsafe URL schemes from passthrough HTML. - Cross-model review gate: Academic-template renders are audited by a fresh Codex thread for rendering fidelity before delivery; dashboard renders skip review by default but accept --review. - Use Case: After an overnight auto-review run, render review-stage/AUTO_REVIEW.md with --state REVIEW_STATE.json to get a readable round-by-round score progression page for the human reviewer. ## Quick Start Ask the assistant to render idea-stage/IDEA_REPORT.md into an HTML report using the academic template.

Frequently Asked Questions about render-html

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

FAQPage Schema
How do I render a Markdown report to a single HTML file?▼

Run the render_html.py helper with the input Markdown path, for example python3 render_html.py idea-stage/IDEA_REPORT.md. It writes a self-contained HTML file next to the source using the academic template by default, with no pip dependencies required.

What Markdown features does the renderer support?▼

It supports ATX headings, paragraphs, bold, italic, strikethrough, inline code, links, images, nested lists, fenced code blocks, tables with alignment, emoji-prefixed blockquote callouts, details/summary HTML blocks, and LaTeX math via MathJax. Footnotes, definition lists, reference-style links, and setext headings are intentionally unsupported.

Can I render HTML offline without CDN access?▼

Yes, pass the --offline flag to skip loading MathJax and highlight.js from cdn.jsdelivr.net. Math then appears as raw $x$ text and code blocks lose syntax highlighting, but the page remains fully readable as a single file.

Does the renderer sanitize unsafe HTML in Markdown?▼

Yes, the helper strips script, style, iframe, form, and similar tags, removes on* event-handler attributes, and rewrites javascript:, vbscript:, and data: URLs to a blocked placeholder. Markdown text content is HTML-escaped separately before any sanitization runs.

Why does the academic template trigger a Codex review step?▼

ARIS treats academic-template HTML as a reviewed deliverable, so a fresh cross-family Codex thread audits rendering fidelity, structure, callout routing, and escaping before delivery. Dashboard renders skip this by default, and --no-review disables it for interim artifacts.

When should I not use this HTML renderer?▼

Do not use it for LaTeX paper output, whose final form is PDF, for SKILL.md protocol files, or for forensic review traces. It is meant only for whitelisted artifacts that gain from sticky TOCs, callouts, math, or score progressions.

Related Skills

Explore More Education & Research Skills→
wanshuiyinwanshuiyin
15.6k

research-pipeline

Orchestrates end-to-end autonomous research from idea discovery through experiments to paper PDF.

Community
Advanced
u-ichiu-ichi
276

reviewable-design-doc

Generate and govern reviewable design documents as HTML with feedback ingestion.

Community
Advanced
sjtuytcsjtuytc
27

proof-checker

Audit LaTeX mathematical proofs for logical gaps and generate issue reports.

Community
Advanced
Rabbit-IvanRabbit-Ivan

hai-visual-report

Generates multi-section HTML reports with Mermaid diagrams from ideas, requirements, and reviews.

Community
Intermediate