OCR with python

Extract Chinese and English text from images and scanned PDFs, including documents like invoices and contracts, using PaddleOCR in Python.

Security Scan

The skill's files and instructions are consistent with an OCR tool using PaddleOCR; it does not request unrelated credentials, call external endpoints, or perform suspicious file/credential access.

Purpose & Capability

The name/description (OCR for Chinese/English PDFs and images) matches the provided scripts and README. The code imports PaddleOCR and PyMuPDF (fitz) as described, and implements PDF image extraction and image OCR.

Instruction Scope

SKILL.md and scripts/ocr.py limit actions to extracting images from PDFs, running PaddleOCR on images, printing or writing recognized text, and cleaning up temp files. The instructions do not ask the agent to read unrelated system files, transmit data to external endpoints, or access secrets.

Install Mechanism

There is no automated install spec; SKILL.md advises using pip3 to install paddlepaddle and paddleocr. This is an expected, proportionate dependency installation approach for a Python OCR skill (note: pip will fetch packages from PyPI; paddlepaddle is large and platform-dependent).

Credentials

The skill requests no environment variables, credentials, or config paths. The requested runtime imports (paddleocr, fitz) are appropriate for OCR/PDF extraction and are proportional to the stated purpose.

Persistence & Privilege

Skill does not request persistent system privileges, does not set always:true, and does not modify other skills or system-wide agent settings. It runs as a normal invokable tool without elevated persistence.

Assessment

This skill appears to be a straightforward OCR helper. Before installing or running: (1) review and run the script in an isolated environment/virtualenv because pip will install large native packages (paddlepaddle) that may require specific platform/driver support; (2) confirm you trust the source (homepage unknown); (3) avoid feeding sensitive documents to unknown third-party tools if you cannot verify where processing occurs — this script runs locally, but dependencies could change behavior if malicious packages are installed; (4) consider running on a disposable VM or container and inspect temporary files under /tmp if you have multi-user concerns.

OCR Text Recognition

This skill uses PaddleOCR for text recognition, supporting both Chinese and English.

Quick Start

Basic Usage

Perform OCR recognition directly on image or PDF files:

from paddleocr import PaddleOCR

ocr = PaddleOCR(lang='ch')
result = ocr.predict("file_path.jpg")

Dependency Installation

Install dependencies before first use:

pip3 install paddlepaddle paddleocr

Output Format

Recognition results return JSON containing:

• rec_texts: List of recognized text
• rec_scores: Confidence score for each text

Typical Use Cases

1. PDF Scans: Use PyMuPDF to extract images first, then OCR
2. Image Text Recognition: Perform OCR directly on images
3. Multi-page PDFs: Process page by page

Scripts

Common scripts are located in the scripts/ directory.