Open source
Open tools behind PDF Craft
PDF Craft brings document conversion, translation, and reading into one experience. We publish and maintain the libraries behind its PDF and EPUB workflows, and share our work on long-text knowledge in the open.
Projects we maintain
PDF Craft
An MIT-licensed Python library for recovering text and structure from scanned PDFs and exporting Markdown or EPUB. It supports the PDF conversion workflow available on this site.
EPUB Translator
An MIT-licensed Python library for translating EPUB books while keeping the original text available for bilingual reading. It connects to our online EPUB translation experience.
Wiki Graph
An Apache-licensed CLI for compiling long text into searchable, source-traceable knowledge graphs and summaries. It is a separate project in our wider document-to-knowledge work.
Open libraries, distinct products
These repositories are published under their own licenses. The PDF Craft website, hosted service, and client apps have separate terms; the library licenses do not automatically apply to them.