Convert Markdown (.md) files and README docs into styled PDF documents online for free. Preserve GFM tables, code blocks, and headings in your browser.
The CommonMark compiler runs entirely in your browser — your README never touches a server.
Drop one or many .md files, or paste Markdown directly into the editor. Source stays in browser memory only.
Preview headings, code blocks, tables, and lists live. Add more files, reorder, or clear before exporting.
Click Export and download a print-ready PDF with live text, code highlighting, and an auto-built TOC.
How Utilvo's in-browser CommonMark+GFM compiler compares to cloud MD converters that LaTeX-roundtrip your README, break Mermaid diagrams, and lose code syntax color.
| Compiler Capability | Utilvo In-Browser | Server-Side Cloud Markdown |
|---|---|---|
| AST Compiler Pass | CommonMark tokenizer + GFM extension pass in JS, then block-by-block PDF text operators | LaTeX roundtrip through Pandoc, fonts and code color often lost |
| Code Block Syntax Highlight | Prism tokenization in browser, GitHub Light theme baked into PDF text spans | Code rendered as monospace black, keyword colors dropped |
| Mermaid & Diagram Fences | Mermaid compiled to SVG, rasterized at 2x DPI as embedded XObject | Mermaid fences emitted as raw text or broken placeholders |
| Table of Contents | Auto-built from H1-H3, clickable PDF link annotations to heading destinations | No auto TOC, or static LaTeX section list with no live links |
| Source Confidentiality | Browser-only, no upload, no log retention, no CDN cache | Mandatory upload, README files retained for analytics |
Four engineering articles walking through the entire MD to PDF pipeline — from tokenizer pass to layout break and code highlighting.
The first pass walks the source character stream and emits block-level tokens (paragraph, heading, list, blockquote, code, table, hr) and inline-level tokens (text, emphasis, strong, code-span, link, image). The token list feeds the next AST builder pass.
The token list is collapsed into an abstract syntax tree. GFM extensions add table, task list, strikethrough, and autolink nodes. Frontmatter is parsed as YAML and attached to the document root for title, author, and date metadata.
Fenced code blocks are tokenized through Prism with the language tag from the fence info string. Tokens are color-mapped to the GitHub Light theme, then written to the PDF content stream as styled Tj text spans with monospace font subsets.
A virtualized layout pass measures each AST node, breaks at heading level 1 by default, honors custom :::page-break::: directives, and writes the page in PDF user space with a 1-inch margin. Page numbers are added as a footer on every body page.
A side-by-side map of every CommonMark and GFM element type Utilvo's compiler handles, plus how each is rendered into the final PDF.
All six heading levels are recognized. Each heading is rendered with a size-graded font weight and inserted into the PDF outline tree as a named bookmark, so the reader's sidebar navigation jumps to the section in one click.
Triple-backtick fenced blocks and four-space indented blocks are both supported. The language tag drives Prism tokenization for keywords, strings, comments, and numbers. Output is selectable text, not an image — the code stays copyable.
Pipe syntax tables become real PDF table objects with header row striping. Colon alignment markers (:---, :---:, ---:) drive left, center, and right cell alignment. Column widths are computed from the longest cell in each column.
Ordered, unordered, and nested lists are measured for hanging indents. GFM task list items render with checked or unchecked Unicode boxes in the PDF, matching GitHub's display so your project READMEs look identical in print.
How the lightweight Markdown markup compares against word-processor formats and ISO 32000-2 for technical documentation.
| Capability | Markdown | DOCX | ODT | HTML | RTF | TXT | |
|---|---|---|---|---|---|---|---|
| Format | Plain text + symbols | OOXML (ZIP+XML) | ODF (ZIP+XML) | HTML / XML | Rich Text Format | Plain text only | ISO 32000-2 binary |
| Human Readable Source | Yes | No (binary) | No (binary) | Yes | Yes | Yes | No (binary) |
| Version Control (Git) | Excellent (diffs) | Poor | Poor | OK | OK | Excellent | Poor |
| Print-Ready Layout | Requires renderer | Yes | Yes | Requires CSS print | Yes | No | Yes (native) |
| Best For | READMEs, docs, blogs | Business documents | Open-source office | Web pages, email | Cross-platform text | Plain notes, logs | Final distribution |
In-depth answers on the CommonMark and GFM parsing model, syntax-highlighted code blocks, Mermaid diagram rasterization, table-of-contents generation, embedded image base64 handling, and the page-size strategies Utilvo uses to keep documentation crisp on print-ready sheets.
All three. Utilvo uses a CommonMark-compliant parser with GFM extensions for tables, task lists, strikethrough, and autolinks. MultiMarkdown metadata blocks, footnotes, and definition lists are recognized and rendered to the same PDF layout.
Yes. Utilvo runs the code through a Prism tokenization pass in the browser, then renders the tokenized spans with the GitHub Light theme. Keywords, strings, comments, and numbers keep their distinct colors on the printed PDF page.
Yes. Drag and drop every .md file you want, reorder the cards, then click convert. Each file becomes a chapter in the final PDF, with a top-level heading break between files and a unified page-number footer.
Mermaid blocks are detected by language tag, then compiled to SVG in the browser and rasterized at 2x DPI for print sharpness. The resulting image is embedded as a PDF XObject on its own line, centered with a thin border.
Yes. Heading levels 1-3 are scanned during the Markdown parse, then a clickable TOC is prepended to the PDF. Page numbers are computed during layout and rendered as PDF link annotations pointing to the destination heading.
Both. GFM pipe tables become a real PDF table with proper column widths, computed from the longest cell. Colon-aligned columns (left, right, center) are honored and applied as text-alignment on the cell content.
Yes. The whole pipeline runs inside the browser sandbox using client-side JavaScript. Your Markdown files are never uploaded to a remote server, never logged, and never cached on a CDN. Source files stay on your device.
You choose. Auto-fit sizes the PDF page to a comfortable reading width (640 points by default). A4 and US Letter are also available with standard 1-inch margins. Page breaks are inserted at heading level 1 to start every chapter on a fresh page.
Yes. Markdown text becomes real PDF text operators with embedded font subsets, so every word is selectable, copyable, and accessible to screen readers. Code blocks remain as styled text spans, not as images.
The Markdown is parsed in a streaming fashion, one block at a time, so memory usage stays under 100 MB even for 500 KB documents with thousands of internal links. The PDF is built incrementally with a virtualized layout pass.
Yes. Utilvo exposes a CSS panel where you can override font family, heading colors, code theme, table stripes, and page margins. The custom styles are applied to the rendered HTML before the PDF is generated, with no server round-trip.
Chain Markdown-to-PDF export with merge, compression, OCR, and organization workflows to ship a complete documentation pipeline.
Convert raster JPG photos alongside Markdown documentation into a single unified multi-page PDF.
Convert lossless PNG screenshots and embed them inside Markdown-driven PDF reports.
Convert vector SVG diagrams referenced from your Markdown docs into a vector PDF section.
Bundle the Markdown-exported PDF with design assets, spreadsheets, and email threads into one deliverable.
Shrink the Markdown-based PDF — already small since text is real text — for email-friendly delivery.
Run OCR to add a layered text layer to any scanned pages that were embedded as images in the MD source.
Reorder, rotate, and delete individual Markdown pages after conversion so the PDF reads in your intended order.
Convert the Markdown-rendered PDF into an editable DOCX with format-preserving text and heading hierarchy.
Recover broken cross-references, missing font subsets, or corrupt bookmarks in converted Markdown PDFs.