Markdown Viewer

A markdown viewer renders the plain-text formatting in a .md file — headings, tables, code blocks, links, images — into the HTML a browser can display. If you have received a .md file from a colleague, downloaded a README from GitHub, or copied output from an AI assistant, this is the tool that lets you read it properly without installing an editor.

This viewer parses your markdown using GitHub Flavored Markdown rules: the same specification GitHub uses to render README files. That means GFM tables, task-list checkboxes, fenced code blocks with syntax highlighting, strikethrough, autolinks, and footnotes all render the way you would expect them to on GitHub.

The diagnostics panel

Most markdown viewers show you the rendered output and stop there. This one also tells you what the document contains: word count, character count, heading count broken down by level, link count split into external, anchor, and relative, image references, code block count with detected languages, task-list progress, and estimated reading time. These numbers appear the moment the file renders — no second step required.

Below the stats, the tool audits the heading structure and link health. If a heading skips a level (H1 jumps to H3), the tool flags it with the line number. If two headings share the same text — which breaks anchor links — it names both. If a link points to an internal anchor (#section) that does not match any heading in the document, the issue is reported. Images missing alt text are flagged for accessibility.

Compatibility detection

Not all markdown is the same. A file authored in Obsidian might contain [[wiki-links]] and Dataview queries. A Next.js blog post might use MDX with JSX components. A README might include Mermaid diagram blocks or LaTeX math expressions. This viewer detects which features the document actually uses and reports them in a compatibility bar. When something will not render — wiki-links, scripts, iframes, MDX — the bar says so explicitly, rather than silently dropping it.

What the viewer does not do

This is not an editor. There is no code pane, no split view, and no live-editing surface. The Source toggle shows the raw markdown read-only, so you can inspect it without the tool pretending to be something it is not. If you need to write or edit markdown, VS Code and Typora are purpose-built for that job.

It does not convert markdown to PDF, Word, or HTML as downloadable files either. Browser print (Ctrl+P) produces a clean page, and dedicated conversion tools for other formats are on the roadmap. And it does not sync with GitHub, Dropbox, or any cloud service — there is no account, no login, and nothing stored between visits.

How the rendering works

When you drop a file or paste markdown, the tool loads two libraries from a CDN: marked.js for parsing and DOMPurify for sanitisation. Your markdown is parsed into HTML, passed through DOMPurify to strip any embedded scripts, event handlers, or iframes, and then inserted into the output pane. The file itself is read by JavaScript running in your browser and is never transmitted anywhere.

The CDN requests load the rendering libraries, not your content. Once loaded, you can disconnect from the network and the viewer continues to work — including rendering new files. If the CDN is unreachable, the tool degrades to a basic sanitiser that still strips dangerous content.

Need to work with the URLs inside a markdown file? Extract URLs from Text pulls every link out of any pasted content. Writing an llms.txt file? The LLMs.txt Generator produces standard markdown you can preview here. And the Readability Checker scores the prose in your documentation once the markdown is rendered.