convert-lakebook

convert-lakebook

Convert Yuque .lakebook export files to clean Markdown documents.

What This Skill Does

1. Extracts the .lakebook tar archive
2. Parses $meta.json for document metadata (titles, ordering, dates)
3. Converts each document's HTML body to clean Markdown using html2text
4. Outputs individual .md files named by document title
5. Generates an _index.md with a table of all documents

Prerequisites

pip install beautifulsoup4 html2text

Usage

When the user provides a .lakebook file path, run the conversion script:

python3 ~/Desktop/lakebook-converter/convert.py "<lakebook_file_path>" --output "<output_dir>"

Parameters

• <lakebook_file_path>: Path to the .lakebook file (required)
• <output_dir>: Output directory for Markdown files (optional, defaults to ~/Desktop/lakebook-converter/output/)

Default Behavior

If the user just says /convert-lakebook with no arguments:

1. Ask the user for the .lakebook file path
2. Ask where they want the output (suggest ~/Desktop/lakebook-converter/output/)
3. Run the conversion
4. Report results: how many docs converted, where they are, and show the index

Example Invocations

/convert-lakebook ~/Desktop/export.lakebook
/convert-lakebook ~/Downloads/my-notes.lakebook --output ~/Documents/notes/

Output Format

Each document becomes a Markdown file:

# Document Title

[converted content...]

An _index.md file is generated with:

# Document Index

Total: N documents

| # | Title | Words | Created |
|---|-------|-------|---------|
| 1 | [Title](filename.md) | 1234 | 2024-01-01 |

Error Handling

• If dependencies are missing, prompt the user to install them
• If the file doesn't exist, report the error clearly
• If some documents fail to convert, report them but continue with the rest

Post-Conversion Tips

After conversion, suggest the user can:

1. Use /extract-memory <output_dir> to analyze their writing style
2. Read specific documents with the Read tool
3. Search across documents with Grep