Skill

ocr-scanned-pdf

Use when you have a scanned PDF with no embedded text layer and need to add OCR text

npx claudepluginhub danielrosehill/claude-code-plugins --plugin document-to-markdown

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Add text recognition to image-only PDFs using `ocrmypdf`. This preprocesses scanned documents to make them readable by text-extraction tools.

SKILL.md

Similar Skills

cache-components

139.3k

Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.

cache-components

mcp-builder

124.2k

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

9 files

anthropics-skills-13

Stats

Stars0

Forks0

Last CommitApr 30, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

OCR Scanned PDF

Add text recognition to image-only PDFs using ocrmypdf. This preprocesses scanned documents to make them readable by text-extraction tools.

When to use

Scanned PDF has no text layer (image-only)

Before running pdf-to-markdown on a scanned document

User explicitly requests OCR on a specific file

Inputs to gather

PDF file path (required)

Language code: --lang (optional; defaults to eng)

Output path: --out (optional; defaults to <stem>-ocr.pdf next to source)

Procedure

Validate that the input PDF exists and is readable.

Check if the PDF already has a text layer using pdftotext -layout <pdf> -. If output is substantial, warn the user that OCR may be unnecessary.

Run ocrmypdf with sensible defaults: ocrmypdf --language=<lang> <input-pdf> <output-pdf>, where language defaults to eng if not specified.

Mention available flags to the user if they need them: --force-ocr (OCR even if text layer exists), --redo-ocr (re-OCR if a layer already exists).

Save output to the specified path or <stem>-ocr.pdf next to the source.

Output / side effects

New PDF file with embedded text layer at the specified output path

Console confirmation of OCR completion and output location

Original PDF unchanged

Safety / constraints

OCR is CPU-intensive; warn user that large PDFs may take several minutes

Do not overwrite the source PDF; always create a new file