processing-docx skill
scriptsCreates, edits, and analyzes Microsoft Word documents using various workflows for text extraction, redlining, and OOXML manipulation.
- GitHub stars
- 245
- Files
- 55
- Last commit
- Jul 19, 2026
- License
- MIT
- Allowed tools
- 5
What does the processing-docx skill do?
processing-docx is an agent skill. Creates, edits, and analyzes Microsoft Word documents using various workflows for text extraction, redlining, and OOXML manipulation.
Summary generated from the skill's SKILL.md.
About the processing-docx skill
Managing Microsoft Word documents through automated workflows is what this skill enables for developers working with legal, academic, or business files. It provides specific paths for different document tasks, such as using pandoc to convert files to markdown for quick text extraction or using docx-js to build new documents from scratch using Paragraph and TextRun components.
An agent should load this skill when it needs to manipulate .docx files, specifically when it must perform redlining to track changes in documents created by others. The skill relies on tools like LibreOffice for visual analysis via PDF conversion and Python scripts for unpacking and repacking OOXML structures. It is strictly limited to Word documents and should not be used for spreadsheets, presentations, or plain text files.
What an agent can do with it
- 1Extract text for analysisConvert a Word document into markdown format while including all tracked changes.
- 2Perform legal redliningEdit a document created by another party using a tracked changes workflow.
- 3Analyze document structureUnpack a file to inspect its raw XML, media, and formatting details.
Written from the skill's SKILL.md. Every use case is checked against a quote from it.
How to install the processing-docx skill
Each command copies the skill folder into the client's skills directory. Restart the client or start a new session to load it.Install the processing-docx skill in Plugin
/plugin marketplace add telagod/code-abyss /plugin install code-abyss@code-abyss
Install the processing-docx skill in Claude Code
~/.claude/skills/processing-docx/SKILL.md
$ npx degit telagod/code-abyss/skills/processing-docx ~/.claude/skills/processing-docxInstall the processing-docx skill in Claude Code (project)
.claude/skills/processing-docx/SKILL.md
$ npx degit telagod/code-abyss/skills/processing-docx .claude/skills/processing-docxInstall the processing-docx skill in Codex
~/.agents/skills/processing-docx/SKILL.md
$ npx degit telagod/code-abyss/skills/processing-docx ~/.agents/skills/processing-docxInstall the processing-docx skill in Gemini CLI
~/.gemini/skills/processing-docx/SKILL.md
$ npx degit telagod/code-abyss/skills/processing-docx ~/.gemini/skills/processing-docxInstall the processing-docx skill in Claude.ai
npx degit telagod/code-abyss/skills/processing-docx processing-docx && zip -r processing-docx.zip processing-docx # claude.ai → Customize → Skills → + → Create skill → Upload a skill → processing-docx.zip
Automated checks
Ran Nov 16, 2026Full report →Not verified yet. A skill needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachabletelagod/code-abyss/skills/processing-docx is public and reachablegithub.com/telagod/code-abyss/tree/HEAD/skills/processing-do ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit 2026-07-19github.com/telagod/code-abyss/commits ↗Pass
- No download-and-run or credential theft in scripts5 of 11 scripts read; none of them downloads and runs code, the rest were not readFinding
- No hidden instructions in SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/telagod/code-abyss/blob/HEAD/skills/processing-do ↗Pass
- Pre-approved tools are scopedBash lets the agent run any shell command without askinggithub.com/telagod/code-abyss/blob/HEAD/skills/processing-do ↗Finding
- SKILL.md follows the Agent Skills formatValid name and description; says when to use it; 62 linesgithub.com/telagod/code-abyss/blob/HEAD/skills/processing-do ↗Pass
SKILL.md
View on GitHubDOCX Processing
.docx is a ZIP archive of XML and resources. Different tasks have different tools and workflows.
Workflow Decision
| Intent | Workflow | Reference |
|---|---|---|
| Read/analyze text only | pandoc → markdown | raw-xml-access.md |
| Read structure, comments, media, formatting | unpack → raw XML | raw-xml-access.md |
| Create new document | docx-js (JS/TS) | docx-js.md |
| Edit own document, simple changes | Document library (Python) | ooxml.md |
| Edit someone else's document | Redlining (tracked changes) | redlining.md |
| Legal / academic / business / gov docs | Redlining — REQUIRED | redlining.md |
| Visual analysis | soffice → PDF → pdftoppm | raw-xml-access.md |
Text Extraction (Quick)
pandoc --track-changes=all path-to-file.docx -o output.md
# --track-changes=accept/reject/all
Create New Document
- MANDATORY — READ ENTIRE FILE:
docx-js.md(~500 lines). NEVER set range limits. - Create JS/TS file using Document, Paragraph, TextRun components.
- Export with
Packer.toBuffer().
Edit Existing Document (Own, Simple)
- MANDATORY — READ ENTIRE FILE:
ooxml.md(~600 lines). NEVER set range limits. python ooxml/scripts/unpack.py <office_file> <output_dir>- Run Python script using Document library.
python ooxml/scripts/pack.py <input_dir> <office_file>
Edit Someone Else's Document → Redlining
See redlining.md for full 6-step workflow with batching strategy and RSID preservation.
Code Style
- Write concise code, no verbose variable names, no unnecessary print statements.
Dependencies
| Package | Install | Purpose |
|---|---|---|
| pandoc | apt install pandoc |
Text extraction |
| docx | npm i -g docx |
Create new docs |
| LibreOffice | apt install libreoffice |
PDF conversion |
| Poppler | apt install poppler-utils |
pdftoppm for images |
| defusedxml | pip install defusedxml |
Secure XML parsing |
Files in this skill
55processing-docx skill questions
How do I install the processing-docx skill in Claude Code?
Run npx degit telagod/code-abyss/skills/processing-docx ~/.claude/skills/processing-docx. Claude Code loads skills from ~/.claude/skills (all projects) or .claude/skills (one project) and uses a skill when its description matches the task.
Which clients can use the processing-docx skill?
Any client that reads the Agent Skills format: Claude Code, Codex, Gemini CLI and claude.ai (upload as a ZIP). Scripts in the skill run only where the client allows code execution.
How can I quickly extract text from a .docx file?
You can use pandoc to convert the document to markdown, with options to track changes.
What should I use for editing legal or academic documents?
For legal, academic, business, or government documents, redlining is required.