Markkit
Copyright 2024-2026 Everruns

Licensed under the MIT License.

=== Design Inspirations ===

Markkit is an independent implementation. The following projects
inspired aspects of its design:

* Readability (Apache 2.0) – Mozilla's approach to content extraction
  informed our HTML-to-markdown pipeline design.

* Pandoc (GPL-2.0) – The universal document converter's format
  coverage and conversion philosophy inspired our multi-format
  architecture.

* turndown (MIT) – The JavaScript HTML-to-markdown converter's
  rule-based approach influenced our conversion engine design.

Markkit's parser, converter, and extraction pipeline are original
Rust implementations. This project is not affiliated with, endorsed
by, or derived from any of the above projects.

=== Key Dependencies ===

Markkit relies on several permissively licensed crates:

* scraper (MIT) – HTML parsing and CSS selector matching
* html5ever (MIT/Apache-2.0) – HTML5 parsing
* lol_html (BSD-3-Clause) – Low-latency HTML rewriting
* pdf-extract (MIT) – PDF text extraction
* reqwest (MIT/Apache-2.0) – HTTP client
* tokio (MIT) – Async runtime
* serde / serde_json (MIT/Apache-2.0) – Serialization

Full dependency license details are available via `cargo deny check licenses`.
