Usage Guide

CLI Usage

langreader -i INPUT.md -o OUTPUT.html --src SRC --tgt TGT --provider PROVIDER --model MODEL [options]

Required Arguments

-i, --input: Input Markdown file path.
-o, --output: Output HTML file path.
--provider: Provider name configured in multiai (e.g., openai, anthropic).
--model: Model name (e.g., gpt-4o, claude-3-5-sonnet).

Optional Arguments

--src (default: fr): Source language code (e.g., fr, de, ja).
--tgt (default: en): Target language code for translation.
-v, --verbose: Verbosity level (0=silent, 1=headings, 2=preview, 3=full).
--continue: Resume generation from the existing output file if the process was interrupted (e.g., by API rate limits). The tool saves progress incrementally; this option allows you to restart where it left off without re-translating the beginning of the file.
--js-url: URL for the external JavaScript file.
- Default: https://cdn.jsdelivr.net/gh/sekika/langreader@main/js/langreader.js
--css-url: URL for the external CSS file.
- Default: https://cdn.jsdelivr.net/gh/sekika/langreader@main/css/langreader.css

YAML Frontmatter Support

You can add metadata at the top of your Markdown file using YAML syntax. This is useful for adding descriptions, attribution, or license information.

---
description: French reading practice. Based on the article [Alsace](https://fr.wikipedia.org/wiki/Alsace) (as of 2025-12-16). Licensed under [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).
---
# Alsace

L’ Alsace est une région historique...

The description field supports Markdown links ([text](url)), which will be rendered as HTML links in the generated reader’s header.

Helper Script: wiki_to_md.py

We provide a helper script, wiki_to_md.py, to download Wikipedia articles and format them for langreader. It automatically adds the necessary YAML frontmatter, including the license and last updated date.

See the Tutorial for details.