86maid86maidCommunityยท1 Agent Skills Included

ddddocr

Captcha recognition, object detection, and slide matching

Recognizes text in captcha images, detects objects and text regions, and matches slide puzzle positions automatically. Eliminates manual captcha solving and complex image preprocessing with a lightweight, cross-platform binary that needs no OpenCV. Exposes simple REST and MCP endpoints so scripts and agents can solve verification challenges in one request.
npx skills add 86maid/ddddocr --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install ddddocr?โ–ผ

Run `npx skills add 86maid/ddddocr --all -g -y` in your terminal to install this skill globally.

How to solve captcha images automatically?โ–ผ

Start the local service and send a base64-encoded image to the /ocr endpoint, or run the included ocr.py script with your image path to get the recognized text instantly.

Can ddddocr solve slide puzzle captchas?โ–ผ

Yes. It offers two slide matching algorithms that compare the puzzle piece with the background image and return the exact target position.

Does ddddocr work with AI agents via MCP?โ–ผ

Yes. The server implements the Model Context Protocol, exposing ocr, det, slide_match, and slide_comparison as callable tools for any MCP-compatible agent.

Does ddddocr require OpenCV or heavy dependencies?โ–ผ

No. It is a self-contained Rust binary that runs cross-platform on Windows, Linux, and macOS without OpenCV, and prebuilt releases are available.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’