Extracts text from PDFs and converts everything to Markdown, including tables. Built on pdfplumber and designed for WSL environments. Handles multi-page documents, preserves page structure, and formats tables into proper Markdown syntax. Won't work on scanned PDFs since there's no OCR, and complex layouts get simplified. The workflow is straightforward: point it at a PDF path, it runs a Python script, and spits out structured Markdown with page breaks and table formatting. Most useful when you need to process PDF reports or documentation into a format Claude can actually work with instead of just describing what's in the file.
- skills.sh installs
- 571
- last updated
- Aug 21, 2026
Install
$ npx -y skills add childbamboo/claude-code-marketplace-sample --skill pdf-reader --agent claude-codeInstalls into .claude/skills of the current project.
Files
PDF Reader
PDF ファイルをテキスト抽出して Markdown 形式に変換するスキルです。
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_pdf.py "/mnt/c/path/to/file.pdf"
Markdown形式で保存
- スクリプトでテキスト抽出
- Write ツールで .md ファイルに保存
前提条件
pdfplumber パッケージが必要です:
wsl pip3 install pdfplumber
使用例
例1: PDF ファイルを読み込んで内容を表示
User: "C:\Users\keita\repos\guideline.pdf を読み込んで"
Assistant:
1. Windowsパスを WSL パスに変換: /mnt/c/Users/keita/repos/guideline.pdf
2. wsl python3 scripts/read_pdf.py を実行
3. 抽出されたテキストを Markdown 形式で表示
例2: PDF を Markdown に変換して保存
User: "ガイドライン.pdf を Markdown に変換して保存"
Assistant:
1. scripts/read_pdf.py でテキスト抽出
2. Markdown形式で構造化(ページごとに見出し、テーブルも含む)
3. Write ツールで ガイドライン.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
- ユーザーが PDF ファイルパスを指定
- Windows パスを WSL パス形式に変換 (
C:\→/mnt/c/) wsl python3 scripts/read_pdf.pyを実行- 抽出されたテキストを Markdown 形式で表示または保存
複数ファイルの一括処理
- Glob で .pdf ファイルを検索
- 各ファイルに対してスクリプトを実行
- 結果をまとめて報告
出力形式
Markdown 構造
# [PDFファイル名]
**Total Pages:** 10
---
## Page 1
[ページ1のテキスト内容]
### Tables
**Table 1:**
| 列1 | 列2 | 列3 |
| --- | --- | --- |
| データ1 | データ2 | データ3 |
---
Sponsors
Details
- Repository
- github.com/childbamboo/claude-code-marketplace-sample
- Categories
- Code Review & Quality, Office & Documents
- First seen
- Jun 3, 2026
