Prefer HTML Output Skill — Documentation skill for Claude Code
Claude Code skill: decide when to render output as HTML over Markdown — based on Thariq Shihipar's html-effectiveness examples.
How to install Prefer HTML Output Skill
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Wechat-ggGitHub/prefer-html-output-skill and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Prefer HTML Output Skill does
Claude Code skill: decide when to render output as HTML over Markdown — based on Thariq Shihipar's html-effectiveness examples. Bilingual (zh-CN / en).
Alternatives in Documentation
- Xlsx — Create, edit, and analyze Excel spreadsheets with formulas, formatting, and visualization 94.1k ★
- 会话管理（Session Manager）需求文档（PRD / Markdown） — 目标：对 Codex / Claude Code 的本地会话记录进行可视化管理，并提供“一键复制 / 一键终端恢复”能力 31.7k ★
- Repo Recap — Generate a structured recap of the repository state: open PRs, open issues, recent releases, and executive sum 11.9k ★
README
prefer-html-output
[](./LICENSE) [](https://claude.com/claude-code) [](https://github.com/ThariqS/html-effectiveness) [](#) [](https://github.com/Wechat-ggGitHub/prefer-html-output-skill/stargazers) [](https://github.com/Wechat-ggGitHub/prefer-html-output-skill/pulls)
A Claude Code skill that lets the agent decide when to render output as HTML instead of Markdown — based on [Thariq Shihipar's](https://x.com/trq212) [html-effectiveness](https://github.com/ThariqS/html-effectiveness) 20 examples.
**Languages**: [English](./README.md) · [简体中文](./README.zh-CN.md)
Why this skill exists
Anthropic Claude Code's engineering lead Thariq Shihipar published [html-effectiveness](https://github.com/ThariqS/html-effectiveness) in May 2026 — 20 self-contained HTML files showing scenarios where HTML beats Markdown for LLM output. In a 20-way head-to-head comparison, HTML won 17.
The operational rule is: **HTML when the document has a human reader; Markdown when it'll be consumed by an automated pipeline or collaboratively edited.**
This skill packages that rule into something an agent can actually follow:
- Detects high-value scenarios during reply generation (weekly reports, PR reviews, design comparisons, concept explainers, flowcharts, design systems, ...)
- Proposes the upgrade with a one-sentence prompt referencing the matching example
- Generates the HTML with the matching style, drops it in a sensible folder, and opens the
Related Skills
HTML Effectiveness Skill
Claude Code skill: Generate interactive HTML instead of markdown walls of text. 9 output patterns, zero deps,
Think Render
Render the most recent structured thought output (from /think) as a diagram or document. Usage: /think-render
Extract Examples
Extract code blocks from markdown files and compile them into a comprehensive reference document with table of
Create Patent
Draft a complete patent application from invention disclosure through assembly of a draft filing package (mark
Fec Review
Conduct a standardized review of the specified file or recently changed front-end code, output a graded review
Business Idea
Write a comprehensive markdown analysis doc for a business idea, product pivot, or new product direction. Outp
Related Agents
Huella Reporter
Assembles the bilingual (Spanish) YOUR_ORG "Informe de Huella Digital" for a huella_digital engagement. Reads
Notion Publisher
Mirrors a markdown source-of-truth file to Notion as a child page under the right project parent, with consist
Phase Writer
Render one MVP build-phase spec into a markdown file at the right path. Use this after the orchestrator has de
