A flexible web crawler and scraping tool using Playwright, supporting both BFS and DFS crawling strategies with screenshot capture and structured output. Installable via npm and usable both as a CLI and programmatically.

a fast chrome extension for students, researchers, and power users. it captures text from any web page and sends it to obsidian, notion, and anki. keeps questions and lists in order. no manual copy paste. just grab and save.

Javascript code snippet for auto scroll on Medium Publishers' List Page

A sleek, minimalist Chrome extension to instantly extract web content, download bulk images/PDFs, and process text natively using Google's Gemini AI.

WordPress plugin that automatically extracts content from external URLs and publishes posts via scheduled cron jobs.

A human-in-the-loop Web Agent that never makes the same mistake twice. MemBrowser combines VLM, DOM stripping, and multi-modal memory to turn brittle AI interactions into robust, zero-cost RPA workflows.一个“教一次就会”的混合架构 Web Agent。结合人类示教（Human-in-the-loop）与多模态记忆网络，彻底解决大模型浏览器自动化的幻觉与高昂成本。

Minimal MV3 Chrome extension that drafts concise, AI-assisted LinkedIn connection notes from visible profile content.

Read the *visible* text from kapture-dom snapshots without drowning in `<div>` soup.

React component state extraction

Android client for the NBP Currency Exchange SOAP backend

Browser extension to calculate total duration of Bilibili favorite lists.