← 返回 Skills 市场
441126098

Md To Rich Html

作者 441126098 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
38
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install md-to-rich-html
功能描述
Turn Markdown files or folders into a polished, human-readable, self-contained HTML page by having the agent understand the document, choose an appropriate i...
使用说明 (SKILL.md)

Markdown to Rich HTML

Use this skill when the user wants to render, preview, publish, share, or make Markdown easier to read. It supports a single .md file or a folder of .md files.

This is an agent-first skill. The goal is not to mechanically convert Markdown tags into HTML tags. The goal is to read the source, infer the reader's task, and design a page that makes the content easier to scan, understand, compare, and use.

Default Workflow

  1. Inspect the input file or folder and read the Markdown content.
  2. Determine the document's real reading purpose, such as a report, guide, interview pack, decision record, implementation plan, review, tutorial, knowledge note, changelog, or mixed document.
  3. Identify local semantic blocks instead of relying only on document type:
    • executive summary, key takeaways, risks, decisions, action items
    • steps, timeline, roadmap, checklist, status, owners
    • Q&A, interview questions, prompts, answers, follow-ups
    • comparisons, trade-offs, pros/cons, options, matrices
    • metrics, counts, constraints, assumptions, warnings
    • code examples, API references, diagrams, tables, citations
  4. Design an information structure for the page before writing HTML:
    • useful first viewport with title, context, summary, and document-specific signals
    • navigation or section map when the document is long
    • component choices that match the content, not just generic prose
    • responsive behavior for desktop and mobile
  5. Generate a self-contained HTML file with inline CSS and JS.
  6. Validate the HTML statically for complete structure, self-contained assets, valid links/anchors where practical, responsive CSS, and obvious accessibility issues.
  7. Open a browser for visual inspection only when the user explicitly asks for preview/verification, the page has complex layout, JavaScript interaction, Mermaid, charts, or meaningful responsive behavior, or the agent changed an existing visual design.
  8. Iterate if the result still looks like plain Markdown, has weak hierarchy, crowded spacing, broken wrapping, inaccessible contrast, or mobile layout issues.

Design Rules

  • Do not preserve Markdown structure mechanically when a better reading structure is obvious.
  • Do not invent facts. Summaries, stats, badges, labels, and classifications must be derived from the source content or clearly phrased as structural labels.
  • Prefer a small set of reusable human-reading components over document-type templates:
    • summary strip
    • stat grid
    • callout
    • comparison table or option matrix
    • timeline
    • step flow
    • collapsible Q&A
    • checklist panel
    • decision card
    • risk/action item list
    • code explanation block
    • diagram section
  • Use visual hierarchy deliberately: typography, spacing, borders, muted backgrounds, accent colors, section labels, and stable component dimensions.
  • Avoid generic landing pages. The first screen should expose the actual content.
  • Keep the output portable as a single HTML file unless the user asks for external assets.
  • If the Markdown contains Mermaid, either preserve and load Mermaid or convert the diagram to a readable fallback block.
  • If the Markdown contains code, preserve syntax and context; add captions or explanation panels when helpful.
  • If the content is short and plain, keep the design restrained. Do not over-design simple notes.

Quality Bar

Before finishing, check the generated page against these questions:

  • Can a reader understand the document's purpose within the first viewport?
  • Are the most important points surfaced before the long body text?
  • Are repeated structures, such as questions, options, tasks, or releases, rendered as components rather than paragraphs?
  • Are dense tables, code, and diagrams readable on desktop and mobile?
  • Does the page have enough hierarchy and contrast without becoming decorative noise?
  • Does the output avoid the "Markdown with nicer CSS" problem?

Use static inspection as the default verification method. Browser screenshots and console checks are optional quality gates for complex pages, not a required step for every conversion.

Maintenance

  • Prefer improving this skill's guidance and quality bar before adding new detector code.
  • Add helper scripts only when they provide narrow tooling, such as file collection, asset copying, HTML validation, or screenshot capture.
  • Do not move the core design decision back into mechanical detector or template enumeration.
安全使用建议
This is reasonable to install for converting Markdown into richer HTML. Use it only with Markdown files you are comfortable letting the agent read, and review the generated HTML before publishing or sharing, especially when the source document contains private information or diagrams that may require external rendering libraries.
能力评估
Purpose & Capability
The artifacts consistently describe reading user-selected Markdown files or folders, understanding their structure, and generating a self-contained HTML page; these capabilities are coherent with the stated purpose.
Instruction Scope
Runtime instructions are bounded to Markdown conversion, static validation, and optional browser inspection for complex or user-requested previews; there are no hidden role changes, prompt overrides, or unrelated automation.
Install Mechanism
The package contains only Markdown, YAML metadata, and a license file, with no executable scripts, declared dependencies, API keys, or install-time commands beyond the normal skill installer flow.
Credentials
The skill must read local Markdown content and write HTML output, which is expected for conversion work; users should avoid applying it to documents they do not want an agent to inspect or potentially include in generated output.
Persistence & Privilege
No artifact requests persistence, background workers, privilege escalation, credential/session access, broad indexing, deletion, or remote exfiltration.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install md-to-rich-html
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /md-to-rich-html 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of md-to-rich-html. - Converts single Markdown files or folders into polished, self-contained HTML pages. - Inspects and understands document purpose before designing an adaptive information structure. - Generates custom HTML, CSS, and JS components tailored to content and reading tasks. - Performs static validation by default; browser-based visual checks are done only for complex or user-requested cases. - Focuses on user-friendly layouts, meaningful hierarchy, and responsiveness for both desktop and mobile. - Avoids generic or mechanical Markdown-to-HTML conversion; always aims for clarity and usefulness in reading.
元数据
Slug md-to-rich-html
版本 1.0.0
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Md To Rich Html 是什么?

Turn Markdown files or folders into a polished, human-readable, self-contained HTML page by having the agent understand the document, choose an appropriate i... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 38 次。

如何安装 Md To Rich Html?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install md-to-rich-html」即可一键安装,无需额外配置。

Md To Rich Html 是免费的吗?

是的,Md To Rich Html 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Md To Rich Html 支持哪些平台?

Md To Rich Html 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Md To Rich Html?

由 441126098(@441126098)开发并维护,当前版本 v1.0.0。

💬 留言讨论