A tool for creating, reading, editing, and generating Microsoft Word (.docx) files, including templates and the files' underlying XML. It can turn Markdown or structured text into finished Word documents.

Use it to create documents from scratch, fill reusable templates, convert Markdown to Word, apply Chinese typography settings, or make low-level changes to existing .docx files.

It removes the need to format Word documents manually or edit their internal files by hand. It also helps keep document creation separate from the instructions used to produce it.

docx — Skill for Claude Code and Codex by XMZF-vAI

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add XMZF-vAI/clerkbox --skill docx

Clone the repo

git clone --depth 1 https://github.com/XMZF-vAI/clerkbox

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/xmzf-vai/clerkbox/docx/github.svg)](https://agentmods.dev/skills/xmzf-vai/clerkbox/docx)

Your own site

<a href="https://agentmods.dev/skills/xmzf-vai/clerkbox/docx"><img src="https://agentmods.dev/badge/skills/xmzf-vai/clerkbox/docx/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/xmzf-vai/clerkbox/docx"><img src="https://agentmods.dev/badge/skills/xmzf-vai/clerkbox/docx.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 206 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 7,968 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

high Prompt Injection · line 732
Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00206	$0.07968
Opus 5	$0.00103	$0.03984
Sonnet 5	$0.00041	$0.01594
Haiku 4.5	$0.00021	$0.00797

Measured 9d ago against content hash b09edb414649, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

docx scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

The scan reads SKILL.md. This mod also ships 21 executable files (scripts/__init__.py, scripts/accept_changes.py, scripts/comment.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

resources/preset-skills/docx/SKILL.md · 858 lines

How it starts

The opening of the file, as written. The whole thing — 858 lines — stays where its author put it; the contents beside it link to each section on GitHub.

docx-pro: Complete Word document skill

Single self-contained skill for everything related to .docx files — high-frequency Markdown/template workflows on top, full docx-js generation in the middle, and low-level OOXML patching at the bottom.

A .docx file is a ZIP container holding a tree of XML files conforming to the OOXML standard. This skill works at every layer: composing Markdown and rendering with pandoc, generating from scratch with docx-js, and unpacking the archive to edit XML directly.

⚠️ Golden Rule: Compose first, render second

The user's request/spec is NOT your Markdown input. You must:

Read and understand the user's requirements (content, structure, data).
Compose a clean, publication-ready Markdown file containing ONLY the final document content — no meta-instructions, no "排版要求" sections, no "用表格呈现：" directives, no formatting instructions.
Then feed that composed Markdown to the renderer.

❌ WRONG — piping the spec file directly:

   # The user gave you "需求.md" describing what doc to produce;
   # you must NOT render it directly:
   python scripts/md_to_docx.py 需求.md output.docx        # BAD!

✅ CORRECT — compose then render:

   # 1. Write a NEW .md with clean business content only
   #    (organize headings, tables, prose based on the spec)
   # 2. Then render YOUR composed file
   python scripts/md_to_docx.py content.md output.docx     # GOOD

If the user explicitly says "把这个 Markdown 文件原样转成 Word" (convert this exact Markdown as-is), only then may you skip the composition step.

Self-check before rendering: Open your composed .md and ask: "Would I hand this text directly to a client/boss as the document body?" If it contains anything a reader shouldn't see (instructions, meta-commentary, formatting directives, section labels like "排版要求"), remove it before rendering.

Decision Matrix

Goal	Pathway
User gives a brief/spec → produce a Word doc	§Standard workflow → §Markdown pipeline
User has a finished Markdown file → format as Word	§Markdown pipeline (skip composition)
Fill a reusable template with placeholder values	§Template fill
Chinese-language doc with correct CJK typography	§Chinese typography
Build a fully custom doc from scratch	§Generating from scratch with docx-js
Modify an existing `.docx` file	§Patching existing documents
Inspect / extract text	`pandoc` or §Patching → unpack to browse raw XML
Tracked changes, comments, paragraph-level XML	§XML patterns + §Patching existing documents

Read the full file on GitHub · 858 lines

Files

What ships with it

53 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 858 lines · 206 tokens per session scan A b09edb414649

Subscribe to this mod's changes

docx is a skill published in the GitHub repository XMZF-vAI/clerkbox (7 stars, last pushed 3d ago), licensed Apache-2.0. It adds 206 tokens to every session and 7,968 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Documents Writing and docs document

pdf-processing

Extract text from PDFs, fill forms, and merge documents.

PrefectHQ/fastmcp · 11 tokens

authoring

How to write Stencila Markdown (.smd) and work with executable documents in related flavours (.myst, .qmd). Use when creating or editing .smd files, adding executable code chunks, inline expressions, parameters, figures, tables, math, or document metadata, or when unsure how Stencila Markdown differs from plain…

stencila/stencila · 73 tokens

execution

How Stencila executes documents - kernels, staleness-driven re-execution, execution options, and debugging chunks that error or do not re-run. Use when running stencila execute or render, when a code chunk is not re-running, when choosing a language/kernel, or when reading execution errors.

stencila/stencila · 64 tokens

conversion

Converting documents between formats with stencila convert - lossless vs lossy targets, quantifying conversion losses, decoding options, and ingesting documents by DOI, arXiv, or PubMed Central identifier. Use when converting to or from DOCX, PDF, LaTeX, JATS, IPYNB, MyST, Quarto or other formats, or importing…

stencila/stencila · 81 tokens

publishing

Publishing Stencila documents - rendering to HTML/PDF/DOCX with themes, building and pushing workspace sites, publishing to Ghost or Zenodo, and signing outputs with content credentials. Use when the user wants to publish, deploy a site, apply a theme, produce final outputs, or push to Ghost, Zenodo or Stencila Cloud.