docx

DOCX Editing

Read, edit, and export .docx files. No permanent install needed.

• Read/edit: uv run --with python-docx
• PDF export: uvx docx2pdf (uses MS Word via automation)

Arguments

read <file>                                          # display contents
edit <file> '{"old":"new", ...}' [output]            # replace text
set-cell <file> <table> <row> <col> "value" [output] # set table cell
pdf <file> [output.pdf]                              # export to PDF

Operations

Read Document

uv run --with python-docx python3 ${CLAUDE_PLUGIN_ROOT}/skills/docx/scripts/docx_tool.py read "/path/to/file.docx"

Output: paragraphs with indices/styles, tables with row/col, content controls, headers/footers.

Edit (Replace Text)

Replaces placeholder text while preserving formatting. Handles Word's run-splitting.

uv run --with python-docx python3 ${CLAUDE_PLUGIN_ROOT}/skills/docx/scripts/docx_tool.py edit "/path/to/file.docx" '{"{{NAME}}": "John Doe", "{{DATE}}": "2026-02-09"}' "/path/to/output.docx"

If output path omitted, overwrites the input file.

Set Table Cell

uv run --with python-docx python3 ${CLAUDE_PLUGIN_ROOT}/skills/docx/scripts/docx_tool.py set-cell "/path/to/file.docx" 0 2 3 "New Value" "/path/to/output.docx"

Arguments: <table_index> <row> <col> "value". Zero-indexed.

Export to PDF

Uses docx2pdf which automates MS Word for perfect-quality conversion.

uvx docx2pdf "/path/to/file.docx" "/path/to/output.pdf"

If output path omitted, saves PDF next to the docx with .pdf extension.

Workflow

For filling a form/template:

- [ ] Read document to identify fields and structure
- [ ] Prepare replacements JSON
- [ ] Copy original as backup: cp original.docx original.backup.docx
- [ ] Run edit with replacements
- [ ] Read output to verify changes
- [ ] Export to PDF if needed: uvx docx2pdf

Best Practices

1. Always read first to understand document structure
2. Back up originals before editing
3. Use exact text from read output as replacement keys
4. Verify edits by reading the output file