playbooks

home / skills / masayan1126 / masayan-uni-code-plugins / ocr

ocr skill

/plugins/ocr/skills/ocr

This skill extracts text from image files and copies it to the clipboard, speeding up documentation and data extraction workflows.

npx playbooks add skill masayan1126/masayan-uni-code-plugins --skill ocr

Review the files below or copy the command above to add this skill to your agents.

Files (3)
SKILL.md
579 B
---
name: ocr
description: 画像ファイルからテキストを抽出しクリップボードにコピー。「文字起こし」「OCR」などで使用。
---

# OCR Skill

画像ファイルからテキストを抽出し、クリップボードにコピーするスキル。

「文字起こし」「OCR」などのキーワードで実行。

## ワークフロー

1. 画像ファイルパスを受け取る
2. Readツールで画像を読み取り、テキスト抽出
3. `scripts/copy_to_clipboard.sh`でクリップボードにコピー
4. 抽出テキストを表示

Overview

This skill extracts text from image files and copies the result to the system clipboard. It is designed for quick OCR tasks like transcribing screenshots, scanned documents, or photos of printed text. Run it with keywords such as "OCR" or "transcribe" to initiate the workflow.

How this skill works

The skill accepts a path to an image file, uses a Read tool to perform OCR and extract text, then invokes a shell helper to place the extracted text into the clipboard. Finally, it displays the extracted text so you can verify the result. The flow is fully automated once the image path is provided.

When to use it

  • Transcribing screenshots, scanned pages, or photographed documents into editable text
  • Quickly transferring text found in images into chat, notes, or editors via the clipboard
  • Extracting text from receipts, forms, or labels for data entry or archiving
  • Preparing content for translation, summarization, or further processing

Best practices

  • Provide clear, high-resolution images for more accurate OCR results
  • Crop images to the region containing text to reduce noise and improve extraction
  • Verify the clipboard contents after extraction, especially for critical data
  • Use language-appropriate OCR models or settings when working with non-English text
  • Avoid heavily stylized or curved text; plain printed or typed text yields best results

Example use cases

  • Capture a screenshot of an error message and extract the text to paste into a bug report
  • Scan a printed invoice and copy the vendor and total into a spreadsheet
  • Transcribe paragraphs from a photographed textbook page for study notes
  • Extract text from a photo of a business card to quickly add contact details

FAQ

What image formats are supported?

Common image formats like PNG, JPEG, and TIFF are supported; provide a file path to the image.

How do I get the extracted text into other apps?

The skill copies the extracted text to the system clipboard, so you can paste it into any application.