What problem does it solve? Reviewing a batch of Word documents scattered across OneDrive or local folders for spelling and grammar mistakes is tedious and error-prone, especially when folder names are ambiguous and DOCX files resist plain-text reading. This Skill provides a structured workflow to locate the right folder, extract readable text from office documents, and produce a file-by-file proofreading report. ## Core Features & Use Cases - Folder tree discovery: Recursively searches OneDrive or local file trees to resolve the actual target folder, handling date-like folder name variants such as 6-2-26 or 6.2.26. - Robust text extraction: Extracts text from .docx, .doc, .odt, .rtf, and .txt files, treating DOCX as a ZIP package and falling back to python-docx or raw word/document.xml parsing when standard readers fail. - Two-pass proofreading: Checks spelling, grammar, punctuation, and capitalization first, then clarity issues like duplicated words and inconsistent dates, reporting findings per file without silently modifying documents. - Use Case: A user says "go to the OneDrive folder and proofread all the docs in the 6-2-26 subfolder" — the Skill resolves the real path, inventories the documents, extracts text from each, and returns a per-file list of spelling and grammar fixes. ## Quick Start Ask the assistant to find the target folder in your OneDrive or local tree and proofread every document in it for spelling and grammar errors.
