tomevault-io Markdown to HTML with Pandoc
Problem
Blog platforms like Tistory do not support API-based uploads. Content must be
copy-pasted manually. Obsidian uses local image syntax (![[path/to/image.png]])
which is not standard markdown, so pandoc cannot render it as an <img> tag.
Instead, pandoc leaves ![[...]] as plain text in the HTML output. This is
useful because the text markers show where images need to be manually inserted
in the blog editor.
All other markdown elements (headers, code blocks, lists, bold, links, etc.) are converted to HTML normally.
Prerequisites
pandocinstalled via Homebrew (brew install pandoc)$OBSIDIAN_VAULTenvironment variable set in~/.zshrc:
# Add to ~/.zshrc
export OBSIDIAN_VAULT="your obsidian path"
Use ~ relative paths. Never use absolute paths like /Users/username/....
Usage
Run the bundled script with the Obsidian file path:
bash scripts/md-to-html.sh "$OBSIDIAN_VAULT/my-post.md"
Output goes to ~/Downloads/<filename>.html.
What the Script Does
- Converts markdown to standalone HTML - Runs
pandoc -sto produce a complete HTML document with<html>,<head>,<body>tags (the-s/--standaloneflag ensures the output is a valid HTML file, not a fragment) - Preserves line breaks - Uses
-f markdown+hard_line_breaksso single newlines become<br />tags, matching Obsidian's line break behavior - Preserves image markers - Obsidian
![[...]]syntax stays as plain text, serving as placeholders for manual image insertion - Outputs to ~/Downloads - Ready for copy-paste into blog editor
After Export
- Open the generated HTML in a browser or text editor
- Copy the content and paste into the blog editor (HTML mode)
- Find the
![[...]]text markers and manually upload images at those positions
Script Location
scripts/md-to-html.sh - See script comments for detailed explanation
of each flag and step.
Converted and distributed by TomeVault — claim your Tome and manage your conversions.