Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
childbamboo avatar

Docx Reader

  • 299 installs
  • Updated January 23, 2026
  • childbamboo/claude-code-marketplace-sample

docx-reader is an Office & Documents skill that helps coding agents read and extract text from Microsoft Word .docx files for use in AI and agent-building document workflows.

About

Reads .docx files and extracts their text content, converting to Markdown on request via a python-docx script. Used when a developer needs to read or convert Word documents.

  • Extracts .docx text via python-docx
  • WSL path conversion and Markdown export

Docx Reader by the numbers

  • 299 all-time installs (skills.sh)
  • Ranked #186 of 688 Office & Documents skills by installs in the Skillselion catalog
  • Data as of Jul 28, 2026 (Skillselion catalog sync)
npx skills add https://github.com/childbamboo/claude-code-marketplace-sample --skill docx-reader

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs299
Last updatedJanuary 23, 2026
Repositorychildbamboo/claude-code-marketplace-sample

How do coding agents read Word docx files?

Extract text from Microsoft Word (.docx) files and optionally save it as Markdown, using a python-docx script in WSL.

Who is it for?

Developers running Claude Code agents that must ingest stakeholder specs, briefs, or requirements delivered as .docx attachments.

Skip if: Teams whose documentation pipeline is already entirely Markdown, PDF, or API-based with no Word document inputs.

When should I use this skill?

A .docx file path is provided and the agent needs document text extracted before coding, summarizing, or analysis.

What you get

Extracted .docx text content, parsed document sections, and agent-ready plain-text context from Word files.

  • extracted document text
  • parsed section content

Files

SKILL.mdMarkdownGitHub ↗

DOCX Reader

Microsoft Word (.docx) ファイルをテキスト形式で読み込むスキルです。

クイックスタート

基本的な使い方

# WSL環境でPythonスクリプトを実行
wsl python3 scripts/read_docx.py "/mnt/c/path/to/file.docx"

Markdown形式で保存

1. スクリプトでテキスト抽出 2. Write ツールで .md ファイルに保存

前提条件

python-docx パッケージが必要です:

wsl pip3 install python-docx

使用例

例1: .docx ファイルを読み込んで内容を表示

User: "C:\Users\keita\repos\file.docx を読み込んで"
Assistant:
1. Windowsパスを WSL パスに変換: /mnt/c/Users/keita/repos/file.docx
2. wsl python3 scripts/read_docx.py を実行
3. 抽出されたテキストを表示

例2: .docx を Markdown に変換して保存

User: "申請書.docx を Markdown に変換して保存"
Assistant:
1. scripts/read_docx.py でテキスト抽出
2. Markdown形式で整形
3. Write ツールで 申請書.md に保存
4. 保存完了を報告

ワークフロー

単一ファイルの読み込み

1. ユーザーが .docx ファイルパスを指定 2. Windows パスを WSL パス形式に変換 (C:\/mnt/c/) 3. wsl python3 scripts/read_docx.py を実行 4. 抽出されたテキストを表示または保存

複数ファイルの一括処理

1. Glob で .docx ファイルを検索 2. 各ファイルに対してスクリプトを実行 3. 結果をまとめて報告

スクリプト詳細

Python スクリプトは scripts/read_docx.py に配置されています。

主な機能:

  • 段落テキストの抽出
  • テーブルデータの抽出
  • エラーハンドリング

使い方:

python scripts/read_docx.py <file_path>

制限事項

  • 画像は抽出されません
  • 複雑なレイアウトは簡略化されます
  • フォント情報、色などのスタイルは失われます
  • 埋め込みオブジェクトは抽出されません

トラブルシューティング

python-docx がインストールされていない

wsl pip3 install python-docx

"No module named 'docx'" エラー

wsl pip3 uninstall docx
wsl pip3 install python-docx

ファイルが開けない

  • ファイルパスが正しいか確認(Windows → WSL パス変換)
  • ファイルが他のプログラムで開かれていないか確認
  • ファイルのアクセス権限を確認

パス変換

Windows パスから WSL パスへの変換:

  • C:\Users\.../mnt/c/Users/...
  • D:\Projects\.../mnt/d/Projects/...
  • バックスラッシュ \ をスラッシュ / に変換

関連ツール

  • pandoc: より高度な変換が必要な場合
  • python-docx2txt: 軽量な代替ライブラリ
  • mammoth: HTML形式での変換

バージョン履歴

  • v1.0.0 (2026-01-06): 初期リリース
  • 基本的なテキスト抽出機能
  • テーブル抽出対応
  • WSL環境での動作

Related skills

FAQ

What file type does docx-reader handle?

docx-reader handles Microsoft Word .docx files, extracting readable text so coding agents in the Claude Code marketplace sample can use document content during AI and agent-building tasks.

When should developers invoke docx-reader?

Developers should invoke docx-reader when a stakeholder spec, brief, or requirements document arrives as a .docx file and the agent needs extracted text before continuing coding or analysis.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.