[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-kortix-ai--suna":3,"tool-kortix-ai--suna":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":68,"owner_location":68,"owner_email":68,"owner_twitter":79,"owner_website":80,"owner_url":81,"languages":82,"stars":120,"forks":121,"last_commit_at":122,"license":123,"difficulty_score":10,"env_os":124,"env_gpu":124,"env_ram":124,"env_deps":125,"category_tags":135,"github_topics":136,"view_count":140,"oss_zip_url":68,"oss_zip_packed_at":68,"status":16,"created_at":141,"updated_at":142,"faqs":143,"releases":172},512,"kortix-ai\u002Fsuna","suna",null,"Suna 是一个强大的开源平台，专门用于创建和管理自主运行的 AI 智能体。它让复杂的自动化任务变得简单，无论是研究分析、数据处理还是网页操作，都能由智能体代为完成。Suna 解决了传统脚本难以应对多步骤、跨平台工作流的痛点，让 AI 真正能像员工一样独立行动。\n\n对于开发者来说，Suna 提供了可视化的构建工具和丰富的能力库，包括浏览器自动化、文件系统管理、系统命令执行以及外部 API 集成。它还支持私有化部署，确保数据安全。即使是非技术背景的用户，也能通过其直观的界面定制专属的智能助手。\n\n特别值得一提的是 Suna 内置的“超级工人”通用型智能体，它展示了平台在自然对话中协调多种任务的能力。从市场调研到文档生成，Suna 都能灵活应对。如果你希望释放人力，让 AI 成为你的数字员工，Suna 是实现这一目标的理想选择。","\u003Cdiv align=\"center\">\n\n# Kortix\n\n**The complete platform for creating autonomous AI agents that work for you**\n\nBuild, manage, and train sophisticated AI agents for any use case. Create powerful agents that act autonomously on your behalf.\n\n[![Discord Follow](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_050f0fca6dd4.png)](https:\u002F\u002Fdiscord.com\u002Finvite\u002FRvFhXUdZ9H)\n[![Twitter Follow](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Fkortix)](https:\u002F\u002Fx.com\u002Fkortix)\n[![GitHub Repo stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkortix-ai\u002Fsuna)](https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna)\n[![Issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fkortix-ai\u002Fsuna)](https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Flabels\u002Fbug)\n\n\u003C!-- Keep these links. Translations will automatically update with the README. -->\n[Deutsch](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=de) | \n[Español](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=es) | \n[français](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=fr) | \n[日本語](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ja) | \n[한국어](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ko) | \n[Português](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=pt) | \n[Русский](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ru) | \n[中文](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=zh)\n\n![Kortix Screenshot](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_366c1f51cfa9.png)\n\u003C\u002Fdiv>\n\n\n\n\n## 🌟 What Makes Kortix Special\n\n### 🤖 Includes Kortix Super Worker – Flagship Generalist AI Worker\nMeet Kortix Super Worker, our showcase agent that demonstrates the full power of the Kortix platform. Through natural conversation, Kortix Super Worker handles research, data analysis, browser automation, file management, and complex workflows – showing you what's possible when you build with Kortix.\n\n### 🔧 Build Custom Kortix Super Worker-Type Agents\nCreate your own specialized agents tailored to specific domains, workflows, or business needs. Whether you need agents for customer service, data processing, content creation, or industry-specific tasks, Kortix provides the infrastructure and tools to build, deploy, and scale them.\n\n### 🚀 Complete Platform Capabilities\n- **Browser Automation**: Navigate websites, extract data, fill forms, automate web workflows\n- **File Management**: Create, edit, and organize documents, spreadsheets, presentations, code\n- **Web Intelligence**: Crawling, search capabilities, data extraction and synthesis\n- **System Operations**: Command-line execution, system administration, DevOps tasks\n- **API Integrations**: Connect with external services and automate cross-platform workflows\n- **Agent Builder**: Visual tools to configure, customize, and deploy agents\n\n## 📋 Table of Contents\n\n- [🌟 What Makes Kortix Special](#-what-makes-kortix-special)\n- [🎯 Agent Examples & Use Cases](#-agent-examples--use-cases)\n- [🏗️ Platform Architecture](#️-platform-architecture)\n- [🚀 Quick Start](#-quick-start)\n- [🏠 Self-Hosting](#-self-hosting)\n- [🤝 Contributing](#-contributing)\n- [📄 License](LICENSE)\n\n## 🎯 Agent Examples & Use Cases\n\n### Kortix Super Worker - Your Generalist AI Worker\n\nKortix Super Worker demonstrates the full capabilities of the Kortix platform as a versatile AI worker that can:\n\n**🔍 Research & Analysis**\n- Conduct comprehensive web research across multiple sources\n- Analyze documents, reports, and datasets\n- Synthesize information and create detailed summaries\n- Market research and competitive intelligence\n\n**🌐 Browser Automation**\n- Navigate complex websites and web applications\n- Extract data from multiple pages automatically\n- Fill forms and submit information\n- Automate repetitive web-based workflows\n\n**📁 File & Document Management**\n- Create and edit documents, spreadsheets, presentations\n- Organize and structure file systems\n- Convert between different file formats\n- Generate reports and documentation\n\n**📊 Data Processing & Analysis**\n- Clean and transform datasets from various sources\n- Perform statistical analysis and create visualizations\n- Monitor KPIs and generate insights\n- Integrate data from multiple APIs and databases\n\n**⚙️ System Administration**\n- Execute command-line operations safely\n- Manage system configurations and deployments\n- Automate DevOps workflows\n- Monitor system health and performance\n\n### Build Your Own Specialized Agents\n\nThe Kortix platform enables you to create agents tailored to specific needs:\n\n**🎧 Customer Service Agents**\n- Handle support tickets and FAQ responses\n- Manage user onboarding and training\n- Escalate complex issues to human agents\n- Track customer satisfaction and feedback\n\n**✍️ Content Creation Agents**\n- Generate marketing copy and social media posts\n- Create technical documentation and tutorials\n- Develop educational content and training materials\n- Maintain content calendars and publishing schedules\n\n**📈 Sales & Marketing Agents**\n- Qualify leads and manage CRM systems\n- Schedule meetings and follow up with prospects\n- Create personalized outreach campaigns\n- Generate sales reports and forecasts\n\n**🔬 Research & Development Agents**\n- Conduct academic and scientific research\n- Monitor industry trends and innovations\n- Analyze patents and competitive landscapes\n- Generate research reports and recommendations\n\n**🏭 Industry-Specific Agents**\n- Healthcare: Patient data analysis, appointment scheduling\n- Finance: Risk assessment, compliance monitoring\n- Legal: Document review, case research\n- Education: Curriculum development, student assessment\n\nEach agent can be configured with custom tools, workflows, knowledge bases, and integrations specific to your requirements.\n\n## 🏗️ Platform Architecture\n\n![Architecture Diagram](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_0d78322381d5.png)\n\nKortix consists of four main components that work together to provide a complete AI agent development platform:\n\n### 🔧 Backend API\nPython\u002FFastAPI service that powers the agent platform with REST endpoints, thread management, agent orchestration, and LLM integration with Anthropic, OpenAI, and others via LiteLLM. Includes agent builder tools, workflow management, and extensible tool system.\n\n### 🖥️ Frontend Dashboard\nNext.js\u002FReact application providing a comprehensive agent management interface with chat interfaces, agent configuration dashboards, workflow builders, monitoring tools, and deployment controls.\n\n### 🐳 Agent Runtime\nIsolated Docker execution environments for each agent instance featuring browser automation, code interpreter, file system access, tool integration, security sandboxing, and scalable agent deployment.\n\n### 🗄️ Database & Storage\nSupabase-powered data layer handling authentication, user management, agent configurations, conversation history, file storage, workflow state, analytics, and real-time subscriptions for live agent monitoring.\n\n## 🚀 Quick Start\n\nGet your Kortix platform running in minutes with our automated setup wizard:\n\n### 1️⃣ Clone the Repository\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna.git\ncd suna\n```\n\n### 2️⃣ Run the Setup Wizard\n```bash\npython setup.py\n```\nThe wizard will guide you through configuring all required services with progress saving, so you can resume if interrupted.\n\n### 3️⃣ Manage the Platform\n```bash\npython start.py          # Interactive start\u002Fstop\npython start.py start    # Start all services\npython start.py stop     # Stop all services\npython start.py status   # Show service status\npython start.py restart  # Restart all services\n```\n\nThe service manager automatically detects your setup method (Docker or Manual) and manages services accordingly.\n\n### 📋 Viewing Realtime Logs\n\n**Manual Setup (native processes):**\n```bash\n# View both backend and frontend logs\ntail -f backend.log frontend.log\n\n# View backend only\ntail -f backend.log\n\n# View frontend only\ntail -f frontend.log\n```\n\n**Docker Setup:**\n```bash\n# View all service logs\ndocker compose logs -f\n\n# View specific service\ndocker compose logs -f backend\ndocker compose logs -f frontend\n```\n\n### 4️⃣ Add More API Keys (Optional)\nAfter initial setup, you can run `python setup.py` again to:\n- **Add\u002FUpdate API Keys** - Configure additional LLM providers (Anthropic, OpenAI, Groq, etc.), search APIs (Tavily, Firecrawl, etc.), and other integrations\n- **Clear setup and start fresh** - Remove all configuration and start over\n\nThat's it! Your Kortix platform will be running with Kortix Super Worker ready to assist you. Ty mate\n---\n\n\u003Cdiv align=\"center\">\n\n**Ready to build your first AI agent?** \n\n[Get Started](.\u002Fdocs\u002FSELF-HOSTING.md) • [Join Discord](https:\u002F\u002Fdiscord.com\u002Finvite\u002FRvFhXUdZ9H) • [Follow on Twitter](https:\u002F\u002Fx.com\u002Fkortix)\n\n\u003C\u002Fdiv>\n","\u003Cdiv align=\"center\">\n\n# Kortix\n\n**为您创建自主 AI 代理 (AI Agents) 的完整平台**\n\n为任何用例构建、管理和训练复杂的 AI 代理。创建能够代表您自主行动的强力代理。\n\n[![Discord Follow](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_050f0fca6dd4.png)](https:\u002F\u002Fdiscord.com\u002Finvite\u002FRvFhXUdZ9H)\n[![Twitter Follow](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Fkortix)](https:\u002F\u002Fx.com\u002Fkortix)\n[![GitHub Repo stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkortix-ai\u002Fsuna)](https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna)\n[![Issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fkortix-ai\u002Fsuna)](https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Flabels\u002Fbug)\n\n\u003C!-- Keep these links. Translations will automatically update with the README. -->\n[Deutsch](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=de) | \n[Español](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=es) | \n[français](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=fr) | \n[日本語](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ja) | \n[한국어](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ko) | \n[Português](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=pt) | \n[Русский](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=ru) | \n[中文](https:\u002F\u002Fwww.readme-i18n.com\u002Fkortix-ai\u002Fsuna?lang=zh)\n\n![Kortix Screenshot](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_366c1f51cfa9.png)\n\u003C\u002Fdiv>\n\n\n\n\n## 🌟 什么让 Kortix 与众不同\n\n### 🤖 包含 Kortix Super Worker – 旗舰通用 AI 工作者\n认识 Kortix Super Worker，我们的展示型代理，展示了 Kortix 平台的完整力量。通过自然对话，Kortix Super Worker 处理研究、数据分析、浏览器自动化、文件管理和复杂工作流 (workflow)——向您展示使用 Kortix 构建时可能实现的功能。\n\n### 🔧 构建自定义的 Kortix Super Worker 类型代理\n创建您自己的专用代理，针对特定领域、工作流或业务需求进行定制。无论您需要客服、数据处理、内容创作还是行业特定任务的代理，Kortix 都提供构建、部署和扩展它们的基础设施和工具。\n\n### 🚀 完整的平台能力\n- **浏览器自动化**：浏览网站、提取数据、填写表单、自动化网页工作流\n- **文件管理**：创建、编辑和组织文档、电子表格、演示文稿、代码\n- **Web 智能**：爬取、搜索能力、数据提取和综合\n- **系统操作**：命令行执行、系统管理、DevOps 任务\n- **API 集成**：连接外部服务并自动化跨平台工作流\n- **代理构建器**：配置、自定义和部署代理的可视化工具\n\n## 📋 目录\n\n- [🌟 什么让 Kortix 与众不同](#-what-makes-kortix-special)\n- [🎯 代理示例与用例](#-agent-examples--use-cases)\n- [🏗️ 平台架构](#️-platform-architecture)\n- [🚀 快速开始](#-quick-start)\n- [🏠 自托管](#-self-hosting)\n- [🤝 贡献](#-contributing)\n- [📄 许可证](LICENSE)\n\n## 🎯 代理示例与用例\n\n### Kortix Super Worker – 您的通用 AI 工作者\n\nKortix Super Worker 展示了 Kortix 平台作为多功能 AI 工作者的全部能力，它可以：\n\n**🔍 研究与分析**\n- 在多个来源上进行全面的网络研究\n- 分析文档、报告和数据集\n- 综合信息并创建详细摘要\n- 市场研究和竞争情报\n\n**🌐 浏览器自动化**\n- 导航复杂的网站和 Web 应用程序\n- 自动从多个页面提取数据\n- 填写表单并提交信息\n- 自动化重复的基于 Web 的工作流\n\n**📁 文件与文档管理**\n- 创建和编辑文档、电子表格、演示文稿\n- 组织和结构化文件系统\n- 在不同文件格式之间转换\n- 生成报告和技术文档\n\n**📊 数据处理与分析**\n- 清理和转换来自各种来源的数据集\n- 执行统计分析并创建可视化图表\n- 监控关键绩效指标 (KPIs) 并生成见解\n- 整合来自多个 API 和数据库的数据\n\n**⚙️ 系统管理**\n- 安全地执行命令行操作\n- 管理系统配置和部署\n- 自动化开发运维 (DevOps) 工作流\n- 监控系统健康和性能\n\n### 构建您自己的专用代理\n\nKortix 平台使您能够创建针对特定需求的代理：\n\n**🎧 客户服务代理**\n- 处理支持工单和常见问题解答回复\n- 管理用户入职和培训\n- 将复杂问题升级给人工代理\n- 跟踪客户满意度和反馈\n\n**✍️ 内容创作代理**\n- 生成营销文案和社交媒体帖子\n- 创建技术文档和教程\n- 开发教育内容和培训材料\n- 维护内容日历和发布计划\n\n**📈 销售与营销代理**\n- 筛选潜在客户并管理客户关系管理 (CRM) 系统\n- 安排会议并跟进潜在客户\n- 创建个性化外联活动\n- 生成销售报告和预测\n\n**🔬 研发代理**\n- 进行学术和科学研究\n- 监控行业趋势和创新\n- 分析专利和竞争格局\n- 生成研究报告和建议\n\n**🏭 行业特定代理**\n- 医疗：患者数据分析、预约安排\n- 金融：风险评估、合规监控\n- 法律：文档审查、案例研究\n- 教育：课程开发、学生评估\n\n每个代理都可以根据您的要求配置自定义工具、工作流、知识库和集成。\n\n## 🏗️ 平台架构\n\n![Architecture Diagram](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_readme_0d78322381d5.png)\n\nKortix 由四个主要组件组成，它们协同工作以提供完整的 AI 代理开发平台：\n\n### 🔧 后端 API\nPython\u002FFastAPI 服务，通过 REST 端点、线程管理、代理编排以及通过 LiteLLM 与 Anthropic、OpenAI 等进行的大语言模型 (LLM) 集成来驱动代理平台。包括代理构建工具、工作流管理和可扩展的工具系统。\n\n### 🖥️ 前端仪表板\nNext.js\u002FReact 应用程序，提供全面的代理管理界面，包括聊天界面、代理配置仪表板、工作流构建器、监控工具和部署控制。\n\n### 🐳 代理运行时\n每个代理实例的隔离 Docker 执行环境，具有浏览器自动化、代码解释器、文件系统访问、工具集成、安全沙箱和可扩展的代理部署功能。\n\n### 🗄️ 数据库与存储\n由 Supabase 驱动的数据层，处理身份验证、用户管理、代理配置、对话历史、文件存储、工作流状态、分析以及用于实时代理监控的实时订阅。\n\n## 🚀 快速开始\n\n通过我们的自动化设置向导，在几分钟内启动您的 Kortix 平台：\n\n### 1️⃣ 克隆仓库\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna.git\ncd suna\n```\n\n### 2️⃣ 运行设置向导\n```bash\npython setup.py\n```\n向导将引导您配置所有必需的服务，并保存进度，以便在中断后继续。\n\n### 3️⃣ 管理平台\n```bash\npython start.py          # Interactive start\u002Fstop\npython start.py start    # Start all services\npython start.py stop     # Stop all services\npython start.py status   # Show service status\npython start.py restart  # Restart all services\n```\n\n服务管理器会自动检测您的安装方式（Docker 或手动），并相应地管理服务。\n\n### 📋 查看实时日志\n\n**手动安装（原生进程）：**\n```bash\n# View both backend and frontend logs\ntail -f backend.log frontend.log\n\n# View backend only\ntail -f backend.log\n\n# View frontend only\ntail -f frontend.log\n```\n\n**Docker 安装：**\n```bash\n# View all service logs\ndocker compose logs -f\n\n# View specific service\ndocker compose logs -f backend\ndocker compose logs -f frontend\n```\n\n### 4️⃣ 添加更多 API 密钥（可选）\n完成初始设置后，您可以再次运行 `python setup.py` 来：\n- **添加\u002F更新 API 密钥** - 配置额外的 LLM（大型语言模型）提供商（Anthropic, OpenAI, Groq 等）、搜索 API（Tavily, Firecrawl 等）以及其他集成\n- **清除设置并重新开始** - 删除所有配置并重新开始\n\n就是这样！您的 Kortix 平台正在运行，Kortix Super Worker 已准备好协助您。谢谢啦\n---\n\n\u003Cdiv align=\"center\">\n\n**准备好构建您的第一个 AI 智能体了吗？** \n\n[开始入门](.\u002Fdocs\u002FSELF-HOSTING.md) • [加入 Discord](https:\u002F\u002Fdiscord.com\u002Finvite\u002FRvFhXUdZ9H) • [在 Twitter 上关注](https:\u002F\u002Fx.com\u002Fkortix)\n\n\u003C\u002Fdiv>","# Suna (Kortix) 快速上手指南\n\n## 🌟 简介\nSuna (Kortix) 是一个完整的平台，用于构建、管理和训练复杂的自主 AI 代理。它支持浏览器自动化、文件管理、数据分析及系统操作等功能，并提供可视化的 Agent 构建工具。\n\n## 🛠️ 环境准备\n在开始之前，请确保您的开发环境满足以下要求：\n- **操作系统**: Linux \u002F macOS \u002F Windows (需支持 Git Bash 或 WSL)\n- **编程语言**: Python 3.8+\n- **容器化**: Docker & Docker Compose (推荐)\n- **前端依赖**: Node.js (部分功能可能需要)\n- **外部服务**:\n    - [Supabase](https:\u002F\u002Fsupabase.com\u002F) 账号 (用于数据库和认证)\n    - LLM API Key (如 Anthropic, OpenAI, Groq 等)\n    - 搜索 API Key (可选，如 Tavily, Firecrawl)\n\n> 💡 **国内开发者提示**：由于涉及 GitHub 代码托管、Supabase 数据库及海外大模型 API，建议配置网络加速环境以确保连接稳定。\n\n## 🚀 安装步骤\n\n### 1. 克隆仓库\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna.git\ncd suna\n```\n\n### 2. 运行设置向导\n使用向导自动配置所需服务（支持断点续传）：\n```bash\npython setup.py\n```\n*注：此步骤将引导您配置 Supabase 连接及各类 API Key。*\n\n### 3. 启动平台\n根据需求管理服务状态：\n```bash\npython start.py          # 交互式启动\u002F停止\npython start.py start    # 启动所有服务\npython start.py stop     # 停止所有服务\npython start.py status   # 查看服务状态\n```\n\n## 💻 基本使用\n\n### 1. 访问控制台\n启动成功后，通常可通过浏览器访问本地地址（具体端口见日志输出），进入前端 Dashboard。\n\n### 2. 配置 API 密钥\n若初始未配置，可再次运行设置向导添加更多提供商密钥：\n```bash\npython setup.py\n```\n\n### 3. 与 Kortix Super Worker 交互\n平台默认包含通用型 AI 代理 \"Kortix Super Worker\"。您可以在聊天界面中直接下达指令，例如：\n- 执行网页研究\n- 处理文档数据\n- 自动化浏览器任务\n\n### 4. 查看实时日志\n- **Docker 模式**:\n  ```bash\n  docker compose logs -f\n  ```\n- **手动模式**:\n  ```bash\n  tail -f backend.log frontend.log\n  ```\n\n## 🔗 扩展资源\n- **详细部署文档**: [`.\u002Fdocs\u002FSELF-HOSTING.md`](.\u002Fdocs\u002FSELF-HOSTING.md)\n- **社区支持**: [Discord](https:\u002F\u002Fdiscord.com\u002Finvite\u002FRvFhXUdZ9H)\n- **项目主页**: [GitHub](https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna)","某跨境电商运营团队需要在季度末快速分析 50 家竞争对手的定价策略与新品动态，以便及时调整下月的营销推广方案。\n\n### 没有 suna 时\n- 团队成员需手动登录数十个独立站点，逐个截图记录价格与库存状态，耗费大量人力。\n- 跨平台收集的用户评论难以统一格式，人工分类与情感分析耗时超过两天。\n- 缺乏实时监控手段，往往在发现竞品降价后才被动应对，错失最佳销售时机。\n- 最终报告依赖 Excel 手工合并，数据核对过程极易出现人为失误，影响决策准确性。\n\n### 使用 suna 后\n- suna 自主配置爬虫任务，同步抓取多站点商品详情与历史价格曲线，实现全天候数据采集。\n- 内置 NLP 能力自动提炼评论关键词，生成情感分布图表无需人工干预，准确率显著提升。\n- 设定触发规则后，一旦竞品调价即通过邮件发送预警通知，确保团队第一时间获知变动。\n- 直接输出包含可视化图表的完整 PPT 报告，大幅缩短汇报准备周期，释放员工创造力。\n\nsuna 通过全链路自动化替代繁琐的人工采集，使运营团队能更敏捷地响应市场变化并聚焦高价值决策。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fkortix-ai_suna_2ebb5a7a.png","kortix-ai","Kortix","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fkortix-ai_56747da1.png","","kortix","https:\u002F\u002Fkortix.com\u002F","https:\u002F\u002Fgithub.com\u002Fkortix-ai",[83,87,91,95,99,103,106,110,114,117],{"name":84,"color":85,"percentage":86},"TypeScript","#3178c6",57.8,{"name":88,"color":89,"percentage":90},"Python","#3572A5",32.6,{"name":92,"color":93,"percentage":94},"HTML","#e34c26",5,{"name":96,"color":97,"percentage":98},"PLpgSQL","#336790",4,{"name":100,"color":101,"percentage":102},"CSS","#663399",0.2,{"name":104,"color":105,"percentage":102},"Shell","#89e051",{"name":107,"color":108,"percentage":109},"JavaScript","#f1e05a",0.1,{"name":111,"color":112,"percentage":113},"Dockerfile","#384d54",0,{"name":115,"color":116,"percentage":113},"Kotlin","#A97BFF",{"name":118,"color":119,"percentage":113},"Ruby","#701516",19574,3374,"2026-04-05T08:46:30","NOASSERTION","未说明",{"notes":126,"python":124,"dependencies":127},"支持 Docker 或手动部署模式，首次运行需通过 setup.py 向导配置服务及 API Key。核心功能依赖外部大模型 API（如 Anthropic、OpenAI），本地环境主要用于运行管理界面和代理运行时容器。日志查看命令包含 tail，暗示对类 Unix 系统友好。",[128,129,130,131,132,133,134],"FastAPI","Next.js","React","Docker","Docker Compose","Supabase","LiteLLM",[15,13,26,14],[137,138,139],"ai","ai-agents","llm",54,"2026-03-27T02:49:30.150509","2026-04-06T07:14:21.773148",[144,149,154,159,164,168],{"id":145,"question_zh":146,"answer_zh":147,"source_url":148},2052,"后端启动时报错（Basejump 错误），如何正确配置 Supabase 数据库？","请在终端中进入 `backend` 目录后，依次执行以下 Supabase CLI 命令：\n1. 登录：`supabase login`\n2. 链接项目：`supabase link --project-ref your_project_reference_id`\n3. 推送迁移：`supabase db push`\n完成后，在 Exposed Schema 界面中选择 basejump 即可解决该问题。","https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Fissues\u002F84",{"id":150,"question_zh":151,"answer_zh":152,"source_url":153},2053,"前端显示“无法连接后端服务器”，但后端日志无报错，如何解决？","这通常是因为网络地址配置不匹配。请按以下步骤修改代码：\n1. 在 `setup.py` 中，将 `backend_url` 从 `http:\u002F\u002Flocalhost:8000\u002Fapi` 修改为实际主机 IP（例如 `http:\u002F\u002Fmy_host_ip:8000\u002Fapi`）。\n2. 在 `backend\u002Fapi.py` 中，将前端地址（例如 `http:\u002F\u002Fmy_host_ip:3000`）添加到 `allowed_origins` 列表中。\n3. 修改后重新运行 setup 流程。","https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Fissues\u002F679",{"id":155,"question_zh":156,"answer_zh":157,"source_url":158},2054,"自托管时数据库迁移失败或报错（FK errors），有什么修复方案？","如果在迁移过程中遇到扩展函数相关的错误，需要显式指定 schema 前缀。请将 SQL 语句中的以下内容进行修改：\n- `gen_random_bytes(length)` 改为 `extension.gen_random_bytes(length)`\n- `uuid_generate_v4()` 改为 `extensions.uuid_generate_v4()`\n此修复方案已在相关 Pull Request 中验证有效。","https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Fissues\u002F1741",{"id":160,"question_zh":161,"answer_zh":162,"source_url":163},2055,"前端能正常加载，但 AI 代理无法生成或任务中途停止，如何排查？","建议在一个干净的环境中进行部署以避免依赖冲突。推荐步骤如下：\n1. 使用干净的 x86_64 PC 或云服务器（如 EC2\u002FAzure）。\n2. 安装 Ubuntu 22.04 系统。\n3. 运行 `apt update` 安装必要环境。\n4. 克隆代码后，直接运行 `python setup.py` 进行初始化配置。","https:\u002F\u002Fgithub.com\u002Fkortix-ai\u002Fsuna\u002Fissues\u002F691",{"id":165,"question_zh":166,"answer_zh":167,"source_url":148},2056,"目前 SUNA 项目的自托管部署流程是否过于复杂？有简化方法吗？","是的，最新的自托管设置已经过简化。开发者不再需要手动处理复杂的依赖和基础设施脚本。用户可以直接在项目根目录运行 `python setup.py` 即可完成大部分配置工作，大幅降低了部署门槛。",{"id":169,"question_zh":170,"answer_zh":171,"source_url":158},2057,"配置了 OpenRouter 或 Gemini Key 但模型始终没有返回任何内容，可能是什么原因？","根据社区反馈，此问题常与数据库迁移失败或 Schema 配置错误有关。请优先检查数据库迁移是否成功完成，并确保扩展函数使用了正确的命名空间（如 `extension.gen_random_bytes`）。如果数据库配置无误，再检查 API Key 的有效性和网络连通性。",[173,178,183,188,193,198],{"id":174,"version":175,"summary_zh":176,"released_at":177},101508,"v0.8.26","## Fix OpenCode startup timeout on image build machines\n\nOpenCode was stuck in an infinite restart loop on JustAVPS image build machines because the token wait (60s) plus DB lock wait (10s) exceeded the ServiceManager's port binding timeout (15s), causing SIGTERM before OpenCode could start.\n\n### Changes\n\n- **fix**: Reduce OpenCode token wait from 60s to 10s — tokens arrive instantly via Docker env or within seconds via \u002Fenv API\n- **fix**: Increase ServiceManager START_WAIT_MS from 15s to 30s to cover token wait + DB lock wait + startup time\n\n### How to Update\n\nPull the new Docker image and recreate your container:\n```\ndocker pull kortix\u002Fcomputer:0.8.26\n```\nOr click **Update** in the Kortix sidebar.\n","2026-04-03T16:46:51",{"id":179,"version":180,"summary_zh":181,"released_at":182},101509,"v0.8.25","## Fix sandbox env var injection for cloud proxied services\n\nSeed placeholders in the SecretStore were overwriting Docker env vars (TAVILY_API_URL, REPLICATE_API_URL, SERPER_API_URL) on every boot, breaking web search, image generation, and serper in cloud sandboxes.\n\n### Changes\n\n- **fix**: sync-s6-env: skip empty SecretStore values so seed placeholders don't overwrite Docker env vars injected at container creation\n- **fix**: Remove platform-injected URL vars (TAVILY_API_URL, REPLICATE_API_URL, SERPER_API_URL) from seed-env.json — they are set by the infrastructure, not user-configured\n- **fix**: persistEnv returns success\u002Ffailure boolean and guards against empty sandbox URL to prevent silent failures\n- **fix**: ?onboarding-skip no longer loses skip intent when sandbox URL is not yet available during auth redirect\n\n### How to Update\n\nPull the new Docker image and recreate your container:\n```\ndocker pull kortix\u002Fcomputer:0.8.25\n```\nOr click **Update** in the Kortix sidebar.\n","2026-04-03T11:35:06",{"id":184,"version":185,"summary_zh":186,"released_at":187},101510,"v0.8.24","## Suna repo migration, runtime env vars, deploy infrastructure\n\nMigrated to kortix-ai\u002Fsuna as the canonical repo. Fixed Next.js runtime env var handling, deploy workflows, and CORS configuration for dev.kortix.com and kortix.com.\n\n### Changes\n\n- **feature**: Next.js connection() for runtime env vars — Docker images now read process.env at request time, not build time\n- **feature**: API version in health endpoint read from release.json (baked into image)\n- **feature**: pnpm dev:prod script for local debugging against production database\n- **fix**: Self-hosted onboarding auto-skipped via API token sync\n- **fix**: Account-state silent error logging instead of swallowing failures\n- **fix**: Added tier_50_400 and tier_200_1000 to legacy paid tiers for claim flow\n- **fix**: Removed legacy models array from account-state endpoint\n- **improvement**: Zero-downtime deploy script updated for ~\u002Fsuna path\n- **improvement**: Deploy workflow supports fallback SSH secrets for dev\u002Fprod servers\n\n### How to Update\n\nPull the new Docker image and recreate your container:\n```\ndocker pull kortix\u002Fcomputer:0.8.24\n```\nOr click **Update** in the Kortix sidebar.\n","2026-04-02T08:35:59",{"id":189,"version":190,"summary_zh":191,"released_at":192},101511,"v1","## Release Highlights\r\n\r\nSuna is a powerful, open-source generalist AI Agent platform that enables autonomous operation across a wide range of tasks including information gathering, content creation, software development, data analysis, and problem-solving.\r\n\r\n### Core Capabilities\r\n\r\n- **Sandboxed Environment**: Secure Linux-based execution environment with full terminal access\r\n- **Web Browsing**: Integrated Chromium browser for autonomous web navigation and interaction\r\n- **File Management**: Complete file system operations for creating, reading, modifying, and organizing content\r\n- **Web Search**: Real-time information retrieval via Tavily integration\r\n- **Data Processing**: Parsing, cleaning, and analyzing structured data (JSON, CSV, XML)\r\n- **Deploy & Expose**: Tools for deploying web applications and exposing ports for sharing\r\n- **Self-guided Workflow**: Autonomous task planning and execution via todo.md system\r\n\r\n### Technical Features\r\n\r\n- FastAPI-based backend with structured agent runs and thread management\r\n- Redis-powered communication for real-time streaming responses\r\n- Persistent sandboxes via daytona integration\r\n- Supabase database integration for state management\r\n- Support for multiple LLM providers via model aliasing\r\n- Billing and account management system\r\n\r\n### Agent Architecture\r\n\r\n- XML-based tool calling for structured agent operations\r\n- Context management for optimized token usage\r\n- Configurable reasoning effort and thinking steps\r\n- Streaming response processing\r\n- Task iteration with autonomous continuation\r\n","2025-04-22T20:12:36",{"id":194,"version":195,"summary_zh":196,"released_at":197},101512,"v.0.1.11","- SQLite as primary storage backend for ThreadManager and StateManager \r\n- Improved StateManager (CRUD, persistent unique store ID´s)","2024-11-19T03:09:10",{"id":199,"version":200,"summary_zh":201,"released_at":202},101513,"v0.1.8","## What's Changed\r\n- Added base processor classes for extensible tool handling:\r\n  - ToolParserBase: Abstract base class for parsing LLM responses\r\n  - ToolExecutorBase: Abstract base class for tool execution strategies\r\n  - ResultsAdderBase: Abstract base class for managing results\r\n- Added dual support for OpenAPI and XML tool calling patterns:\r\n  - XML schema decorator for XML-based tool definitions\r\n  - XML-specific processors for parsing and execution\r\n  - Standard processors for OpenAPI function calling\r\n- Enhanced streaming capabilities:\r\n  - execute_tools_on_stream: Execute tools in real-time during streaming\r\n","2024-11-18T07:14:52"]