Browser Harness | Self-healing harness that enables LLMs to complete any task.

Running a big model on a small laptop

An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

🎬 火宝短剧 - 基于AI的一站式短剧生成平台 《一句话生成完整短剧，从剧本到成片全自动化》 Huobao Drama - An AI-Powered End-to-End Short Drama Generator "One Sentence to Complete Drama: Fully Automated from Script to Final Video"

💫 Toolkit to help you get started with Spec-Driven Development

百聆 是一个类似GPT-4o的语音对话机器人，通过ASR+LLM+TTS实现，集成DeepSeek R1等优秀大模型，接入openClaw，真正的个人语音助手，时延低至800ms，Mac等低配置也可运行，支持打断

Deep Convolutional Q learning based Self learning implementation for Subway Surfers game

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without …

SafeLine is a self-hosted WAF(Web Application Firewall) / reverse proxy to protect your web apps from attacks and exploits.

Ingest, parse, and optimize any data format ➡️ from documents to multimedia ➡️ for enhanced compatibility with GenAI frameworks

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

低代码框架，支持流式布局与自由布局拖拽编排，可视化拖拽、随意嵌套组合、实时渲染、实时辅助线展示、自由布局支持辅助对齐、支持自动吸附、实时组件间距展示、实时拖拽排序、状态域管理，可视化属性配置、可视化样式配置、多设备适配展示，支持逻辑渲染、模板字符变量、表达式、自定义方法、自定义状态

Code for fintune ChatGLM-6b using low-rank adaptation (LoRA)

Real time streaming talking head

A Modern E-Commerce-Store for Custom Phone Cases

基于AI的图片/视频硬字幕去除、文本水印去除，无损分辨率生成去字幕、去水印后的图片/视频文件。无需申请第三方API，本地实现。AI-based tool for removing hard-coded subtitles and text-like watermarks from videos or Pictures.

Window management made elegant.

Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key

这是一个 Flutter Packge 以实现摄像头精确追踪并识别十指的运动路径/轨迹和手势动作, 且输出22个手部关键点以支持更多手势自定义. 基于这个包可以编写业务逻辑将手势信息实时转化为指令信息: 一二三四五, rock, spiderman...还可以对不同手势编写不同特效. 可用于短视频直播特效, 智能硬件等领域, 为人机互动带来更自然丰富的体验

Video players (IJKplayer, ExoPlayer, MediaPlayer), HTTPS, 16k page size, danmaku (bullet chat) support, external subtitles, support for filters, watermarks, and GIF screenshots, pre-roll and mid-ro…