Provides detailed workflows and best-practice patterns for reading, creating, and editing Word (.docx) documents. Covers text extraction, raw OOXML access, tracked-change redlining, batch editing strategies, conversion to PDF/images, and tooling recommendations (pandoc, docx-js, LibreOffice).
Use this skill when you need to programmatically generate Word documents, implement precise tracked changes in a review workflow, convert documents to markdown or images, or inspect raw XML for comments and embedded media. Especially relevant for legal, academic, or business documents where redlines must be exact.
Agents capable of running shell commands and Python/Node workflows or providing code snippets for document manipulation (Claude Code, Copilot, Cursor).
Comprehensive DOCX skill covering creation (via docx-js), editing (via OOXML/Python), tracked changes/redlining workflow, and document-to-image conversion. No bundled scripts — purely instructional SKILL.md referencing external scripts (unpack.py, pack.py) and reference docs (docx-js.md, ooxml.md) not included. Well-structured decision tree and good BAD/GOOD code examples for tracked changes patterns.
No scripts to test. Skill is purely instructional. The referenced scripts (ooxml/scripts/unpack.py, ooxml/scripts/pack.py) and reference docs (docx-js.md, ooxml.md) are not bundled — would need the full repo for those. Output path rule (/shared/ only) is a nice safety touch. Proprietary license noted in frontmatter.
