inka-clean

Remove inka2 metadata from markdown files while preserving content and adjusting heading hierarchy.

Overview

inka-clean is a Unix pipeline tool that processes markdown files containing inka2 flashcard sections. It removes inka2 metadata (delimiters, deck/tag specifications, ID comments, question numbers, answer prefixes) while preserving the actual content and intelligently adjusting heading levels to fit the surrounding markdown hierarchy.

Installation

uv tool install inka-clean
# or for development
cd inka-clean && make install

Usage

As a Unix pipeline tool (stdin → stdout):

# Read from stdin, write to stdout
cat file.md | inka-clean > output.md

# Direct file redirection
inka-clean < input.md > output.md

# With file argument
inka-clean input.md > output.md

# Chain with other tools
cat file.md | inka-clean | pandoc -o output.html

Example

Input:

## Outer Heading
Some content.

---
Deck: tools::vim
Tags: vim

<!--ID:1755681272665-->
1. How to format text paragraphs?
> `gq`
> `gqap` (current paragraph)

---

### Another Heading
More content.

Output:

## Outer Heading
Some content.

How to format text paragraphs?
`gq`
`gqap` (current paragraph)

### Another Heading
More content.

Features

• Metadata removal: Strips --- delimiters, Deck:, Tags:, <!--ID:--> comments
• Prefix stripping: Removes 1. question numbers and > answer prefixes
• Heading adjustment: Adjusts heading levels in inka2 content relative to surrounding context
• Edge case handling: Preserves code blocks, regular ordered lists, blockquotes, thematic breaks
• Performance: < 20ms for typical files
• Unix philosophy: Pure stdin/stdout, composable with other tools

Development

# Run tests (TDD)
make test

# Type check
make ty

# Lint and format
make lint
make format

# Run benchmarks
make bench

License

MIT

inka-clean