document-diff
Installation
SKILL.md
Document Diff
Overview
Compare two versions of a document with structure-aware precision. SoMark parses both files into clean Markdown first, then a diff is generated at the text level. The result tells you exactly what changed between two versions of a contract, report, policy document, or any other file.
Why parse before diffing?
Raw PDF/Word binary diffing is meaningless. By parsing both documents into clean Markdown first, the diff captures semantic changes — actual content additions, deletions, and modifications — not binary noise.
In short: parse both documents with SoMark, then diff the structured output.