This project provides an AI-powered "skill" for automating the revision and formatting of thesis and dissertation Word documents. It targets students and researchers needing to adhere to strict academic formatting guidelines, offering a structured approach to ensure consistency, fix common document issues, and generate supporting visuals and code snippets, ultimately streamlining the final document preparation process.

How It Works

The tool employs a methodical, AI-driven workflow prioritizing stability and accuracy. It begins by verifying the Word automation environment, followed by a deep audit of the OOXML structure to uncover hidden formatting inconsistencies. Formatting changes are applied in a defined order: first, strictly adhering to institutional requirements, then preserving existing styles where no specific rules apply. The process culminates in exporting a PDF for meticulous page-by-page review, ensuring all modifications are validated.

Quick Start & Requirements

• Installation: Clone the repository: git clone https://github.com/the-shy123456/thesis-docx.git. No further file extraction is needed.
• Prerequisites:
  • Windows operating system with desktop Microsoft Word installed.
  • Python environment with python-docx and lxml libraries.
  • Node.js with mmdc or npx for Mermaid diagram rendering.
• Quick Start Commands:
  • Check Word COM/DOM availability: powershell -ExecutionPolicy Bypass -File scripts/check_word_com.ps1 -Json
  • Audit OOXML: python scripts/audit_docx_ooxml.py .\draft.docx --output_json .\draft.audit.json --output_txt .\draft.audit.txt
  • Normalize styles (dry-run): powershell -ExecutionPolicy Bypass -File scripts/normalize_word_styles.ps1 -InputPath .\draft.docx -AuditOnly
  • Export PDF: powershell -ExecutionPolicy Bypass -File scripts/export_word_pdf.ps1 -DocPath .\draft.docx -PdfPath .\draft.audit.pdf
• Documentation: Links to SKILL.md, README_EN.md, references/script-usage.md, references/paper-format-workflow.md.

Highlighted Details

• Automates revision of thesis/dissertation Word documents, including unified styling for text, headings, captions, and references.
• Fixes common document issues such as table of contents, page numbering, section breaks, cross-references, and figure/table numbering.
• Generates Mermaid diagrams based on real data and produces LaTeX-formatted code snippets or pseudocode.
• Audits OOXML for subtle, hidden problems like styleId, firstLineChars, titlePg, REF field display values, and section-level headers/footers.

Maintenance & Community

The project is hosted on GitHub and mentions the "LINUX DO Community". No specific details on active contributors, sponsorships, or roadmap are provided in the README.

Licensing & Compatibility

The project is released under the MIT License. This license is permissive and generally compatible with commercial use and closed-source linking.

Limitations & Caveats

The tool is explicitly designed for Windows environments with desktop Microsoft Word. It emphasizes a page-by-page PDF review as essential for confirming completion, implying that automated export alone should not be considered a final validation step. The core functionality relies on the stability and accessibility of Word's COM/DOM interface.