ocr-local for Openclaw

A powerful, 100% local OCR tool for extracting Chinese and English text from images without API keys.

shaw555
v1.0.0
Mar 15, 2026
20
20.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install ocr-local

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install ocr-local using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is ocr-local?

The ocr-local skill provides a high-performance, privacy-centric solution for optical character recognition within the Openclaw Skills framework. Built on Tesseract.js, it eliminates the need for expensive or privacy-invasive cloud APIs by performing all image analysis directly on your local machine. This ensures that sensitive data remains secure and reduces operational costs for text extraction tasks.

Supporting Simplified Chinese, Traditional Chinese, and English, this skill is designed for developers who need to bridge the gap between visual assets and textual data. Whether you are building an automated document indexer or an AI agent that needs to read screenshots, ocr-local delivers reliable results with minimal configuration.

ocr-local Use Cases

  • Automating text extraction from system screenshots or user-uploaded images.
  • Processing sensitive financial or legal documents without external data transmission.
  • Creating searchable databases from collections of historical scanned documents.
  • Integrating multi-language OCR capabilities into local AI coding agent workflows.

How ocr-local Works

  1. The user or agent triggers the script by providing an image path and optional language parameters.
  2. The skill checks for the presence of Node.js and the Tesseract.js library.
  3. Upon the first execution, the required language training files are downloaded once and stored in a local cache.
  4. The engine processes the image pixels to recognize characters based on the specified language model.
  5. The final extracted text is formatted and output to the console or saved as a JSON object for further automation.

ocr-local Setup

To get started with this skill, ensure Node.js is installed on your system. Then, install the Tesseract dependency using npm:

npm install tesseract.js

You can then run the OCR script directly from the command line using the following syntax:

node scripts/ocr.js /path/to/your/image.jpg --lang chi_sim+eng

ocr-local Data Schema & Taxonomy

The skill processes image files and produces either raw strings or structured JSON.

Component Details
Supported Formats JPG, PNG
Language Packs eng, chi_sim, chi_tra
Storage Local cache for training data (~20MB per language)
Output Modes Plain Text (default) or Structured Data (--json)

ocr-local Advanced Features

  • Zero-dependency on external APIs ensures 100% uptime and privacy for Openclaw Skills users.
  • Combined language recognition allowing for mixed-script document processing such as Simplified Chinese and English simultaneously.
  • Local data caching which speeds up subsequent recognition tasks significantly after the initial download.
  • Flexible output formatting that facilitates easy ingestion by other automation tools in the Openclaw Skills library.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Bins node
Github Stars: 0
forks: 0

Featured*