数据来源：https://github.com/trending 抓取时间：2026-07-02 09:00 (Asia/Shanghai)
| # | 项目 | 描述 | 语言 | ⭐ Stars | 📈 今日新增 |
|---|---|---|---|---|---|
| 1 | hasaneyldrm/exercises-dataset | 包含 433 个健身动作的综合数据集。每个条目包含名称、类别、目标肌群、所需器械、说明、缩略图和动画视频。<br/>A comprehensive dataset of 433 fitness exercises. Each entry includes name, category, target muscle group, equipment, instructions, thumbnail image, and animation video. | HTML | 8,422 | +2,470 |
| 2 | usestrix/strix | 开源的 AI 渗透测试工具，用于发现和修复应用漏洞。<br/>Open-source AI penetration testing tool to find and fix your app's vulnerabilities. | Python | 29,725 | +1,211 |
| 3 | microsoft/AI-For-Beginners | 12 周，24 节课，适合所有人的 AI 入门课程！<br/>12 Weeks, 24 Lessons, AI for All! | Jupyter Notebook | 50,443 | +1,096 |
| 4 | diegosouzapw/OmniRoute | 免费 AI 网关：一个端点，231+ 提供商（50+ 免费），连接 Claude Code、Codex、Cursor、Cline 和 Copilot。RTK+Caveman 堆叠压缩节省 15-95% tokens，智能自动回退，支持 MCP/A2A、多模态 API，桌面/PWA。<br/>Free AI gateway: one endpoint, 231+ providers (50+ free), connect Claude Code, Codex, Cursor, Cline & Copilot. RTK+Caveman stacked compression saves 15-95% tokens, smart auto-fallback, MCP/A2A, multimodal APIs, Desktop/PWA. | TypeScript | 9,527 | +1,010 |
| 5 | facebook/astryx | 开源设计系统，完全可定制且支持 Agent。<br/>An open source design system that's fully customizable and agent ready. | TypeScript | 2,641 | +708 |
| 6 | ogulcancelik/herdr | 运行在终端中的 Agent 多路复用器。<br/>Agent multiplexer that lives in your terminal. | Rust | 9,602 | +609 |
| 7 | altic-dev/FluidVoice | 最快且唯一的 macOS 听写应用，具备设备端 STT 和自定义训练 AI 增强模型 —— 本地 Wispr Flow 替代方案。即将支持 Windows、iOS 和 Linux。<br/>Fastest and only macOS Dictation app with on-device STT and custom trained AI enhancement model - Local Wispr Flow alternative. | Swift | 5,494 | +572 |
| 8 | CoreBunch/Instatic | 现代自托管可视化 CMS —— 1 分钟即可启动。<br/>Instatic is a modern self-hosted visual CMS - get it running in 1 minute. | TypeScript | 2,012 | +508 |
| 9 | Mebus/cupp | 通用用户密码分析器 (CUPP)。<br/>Common User Passwords Profiler (CUPP). | Python | 6,245 | +184 |
| 10 | 0xNyk/council-of-high-intelligence | 18 个 AI 角色在多个 LLM 提供商之间 deliberates 你最难的决策。亚里士多德、费曼、卡尼曼、托瓦兹等人 —— 结构化多轮 deliberation，真正的模型多样性。一条命令：/council<br/>18 AI personas deliberate your hardest decisions across multiple LLM providers. Aristotle, Feynman, Kahneman, Torvalds & more. | Shell | 2,627 | +161 |
| 11 | refactoringhq/tolaria | 管理 Markdown 知识库的桌面应用。<br/>Desktop app to manage markdown knowledge bases. | TypeScript | 18,018 | +150 |
| 12 | Unclecheng-li/VulnClaw | 基于 AI Agent + MCP 工具链 + 渗透 Skill 编排，配合大语言模型，自然语言输入 → 自动完成「信息收集 → 漏洞发现 → 漏洞利用 → 报告生成」全流程。<br/>AI Agent + MCP toolchain + penetration skill orchestration, with LLM. Natural language input → automated full pipeline. | Python | 1,571 | +132 |
Generated by GitHub Trending Daily Bot · 2026-07-02