tesseract-ocr
by whalefell
Image & Video GenerationBrief
Extract text from images using the Tesseract OCR engine directly via command line.
Description
What This Skill Does
Extracts text from images using the Tesseract OCR engine via command line. Supports English, Chinese, Japanese, Korean, and other languages through installable language packs. No Python or additional runtime dependencies beyond the tesseract binary.
Runs directly from the command line without Python or additional runtimes, making it easy to drop into shell scripts and CI pipelines.
When to Use It
- Extracting text from scanned receipts or invoices
- Reading text content from screenshots
- Processing documents with mixed Chinese and English
- Automating text extraction in shell scripts
- Converting image-based content to plain text files
Example Workflow
User asks: Extract all text from document.png, which contains mixed Chinese and English
- 1
Identify the image path and required language codes
- 2
Run tesseract document.png stdout -l chi_sim+eng
- 3
Capture the text output from stdout
- 4
Return the extracted text to the user
Plain text with both Chinese and English content recognized from the image
Requirements
- tesseract-ocr system package (brew install tesseract on macOS, apt-get install tesseract-ocr on Ubuntu/Debian)
- Language pack for each non-English language needed (e.g., tesseract-ocr-chi-sim for Simplified Chinese, brew install tesseract-lang on macOS)
Installation
clawhub install whalefell/tesseract-ocrRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.