[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-ValueCell-ai--ClawX":3,"tool-ValueCell-ai--ClawX":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":79,"owner_location":79,"owner_email":80,"owner_twitter":81,"owner_website":82,"owner_url":83,"languages":84,"stars":118,"forks":119,"last_commit_at":120,"license":121,"difficulty_score":23,"env_os":122,"env_gpu":123,"env_ram":124,"env_deps":125,"category_tags":132,"github_topics":133,"view_count":142,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":143,"updated_at":144,"faqs":145,"releases":186},274,"ValueCell-ai\u002FClawX","ClawX","ClawX is a desktop app that provides a graphical interface for OpenClaw AI agents. It turns CLI-based AI orchestration into a desktop experience without using the terminal. China website is https:\u002F\u002Fclawx.com.cn.","ClawX 是一款面向 OpenClaw AI 智能体的桌面应用，它将原本依赖命令行的操作转化为直观的图形界面，让用户无需打开终端即可轻松管理和使用 AI 代理。许多 AI 工具虽然功能强大，但配置复杂、依赖命令行操作，对非技术用户门槛较高。ClawX 正是为解决这一问题而生——通过可视化设置、一键安装向导和内置运行时环境，大幅降低使用门槛。\n\n它适合希望高效利用 AI 自动化工作流的普通用户、设计师、产品经理，也兼顾开发者对高级配置的需求（如通过“开发者模式”调整细节）。ClawX 内嵌官方 OpenClaw 核心，确保与上游生态同步，同时支持多平台（Windows、macOS、Linux）、多语言，并提供统一的模型提供商管理、技能插件市场和任务调度界面。整个体验开箱即用，真正实现“强大能力，简单使用”。","\n\u003Cp align=\"center\">\n  \u003Cimg src=\"src\u002Fassets\u002Flogo.svg\" width=\"128\" height=\"128\" alt=\"ClawX Logo\" \u002F>\n\u003C\u002Fp>\n\n\u003Ch1 align=\"center\">ClawX\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>The Desktop Interface for OpenClaw AI Agents\u003C\u002Fstrong>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"#features\">Features\u003C\u002Fa> •\n  \u003Ca href=\"#why-clawx\">Why ClawX\u003C\u002Fa> •\n  \u003Ca href=\"#getting-started\">Getting Started\u003C\u002Fa> •\n  \u003Ca href=\"#architecture\">Architecture\u003C\u002Fa> •\n  \u003Ca href=\"#development\">Development\u003C\u002Fa> •\n  \u003Ca href=\"#contributing\">Contributing\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fplatform-MacOS%20%7C%20Windows%20%7C%20Linux-blue\" alt=\"Platform\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Felectron-40+-47848F?logo=electron\" alt=\"Electron\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Freact-19-61DAFB?logo=react\" alt=\"React\" \u002F>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.com\u002Finvite\u002F84Kex3GGAh\" target=\"_blank\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1399603591471435907?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb\" alt=\"chat on Discord\" \u002F>\n  \u003C\u002Fa>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdownloads\u002FValueCell-ai\u002FClawX\u002Ftotal?color=%23027DEB\" alt=\"Downloads\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-green\" alt=\"License\" \u002F>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  English | \u003Ca href=\"README.zh-CN.md\">简体中文\u003C\u002Fa> | \u003Ca href=\"README.ja-JP.md\">日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\n---\n\n## Overview\n\n**ClawX** bridges the gap between powerful AI agents and everyday users. Built on top of [OpenClaw](https:\u002F\u002Fgithub.com\u002FOpenClaw), it transforms command-line AI orchestration into an accessible, beautiful desktop experience—no terminal required.\n\nWhether you're automating workflows, managing AI-powered channels, or scheduling intelligent tasks, ClawX provides the interface you need to harness AI agents effectively.\n\nClawX comes pre-configured with best-practice model providers and natively supports Windows as well as multi-language settings. Of course, you can also fine-tune advanced configurations via **Settings → Advanced → Developer Mode**.\n\n---\n## Screenshot\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_56a89596f02d.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_48e6d99132e7.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_3e576ab04cac.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_0cbe90d9b324.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_e51903d71a7e.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_46cf0b0d3c0f.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n---\n\n## Why ClawX\n\nBuilding AI agents shouldn't require mastering the command line. ClawX was designed with a simple philosophy: **powerful technology deserves an interface that respects your time.**\n\n| Challenge | ClawX Solution |\n|-----------|----------------|\n| Complex CLI setup | One-click installation with guided setup wizard |\n| Configuration files | Visual settings with real-time validation |\n| Process management | Automatic gateway lifecycle management |\n| Multiple AI providers | Unified provider configuration panel |\n| Skill\u002Fplugin installation | Built-in skill marketplace and management |\n\n### OpenClaw Inside\n\nClawX is built directly upon the official **OpenClaw** core. Instead of requiring a separate installation, we embed the runtime within the application to provide a seamless \"battery-included\" experience.\n\nWe are committed to maintaining strict alignment with the upstream OpenClaw project, ensuring that you always have access to the latest capabilities, stability improvements, and ecosystem compatibility provided by the official releases.\n\n---\n\n## Features\n\n### 🎯 Zero Configuration Barrier\nComplete the entire setup—from installation to your first AI interaction—through an intuitive graphical interface. No terminal commands, no YAML files, no environment variable hunting.\n\n### 💬 Intelligent Chat Interface\nCommunicate with AI agents through a modern chat experience. Support for multiple conversation contexts, message history, rich content rendering with Markdown, and direct `@agent` routing in the main composer for multi-agent setups.\nWhen you target another agent with `@agent`, ClawX switches into that agent's own conversation context directly instead of relaying through the default agent. Agent workspaces stay separate by default, and stronger isolation depends on OpenClaw sandbox settings.\nEach agent can also override its own `provider\u002Fmodel` runtime setting; agents without overrides continue inheriting the global default model.\n\n### 📡 Multi-Channel Management\nConfigure and monitor multiple AI channels simultaneously. Each channel operates independently, allowing you to run specialized agents for different tasks.\nEach channel now supports multiple accounts, per-account agent binding, and switching the channel default account directly from the Channels page.\nClawX now also bundles Tencent's official personal WeChat channel plugin, so you can link WeChat directly from the Channels page with an in-app QR flow.\n\n### ⏰ Cron-Based Automation\nSchedule AI tasks to run automatically. Define triggers, set intervals, and let your AI agents work around the clock without manual intervention.\nThe Cron page now lets you configure external delivery directly in the task form with separate sender-account and recipient-target selectors. For supported channels, recipient targets are discovered automatically from channel directories or known session history, so you no longer need to edit `jobs.json` by hand.\nKnown limitation: WeChat is intentionally excluded from supported cron delivery channels for now. The current `openclaw-weixin` plugin requires a live conversation `contextToken` for outbound sends, so cron-style proactive delivery is not supported by the plugin itself.\n\n### 🧩 Extensible Skill System\nExtend your AI agents with pre-built skills. Browse, install, and manage skills through the integrated skill panel—no package managers required.\nClawX also pre-bundles full document-processing skills (`pdf`, `xlsx`, `docx`, `pptx`), deploys them automatically to the managed skills directory (default `~\u002F.openclaw\u002Fskills`) on startup, and enables them by default on first install. Additional bundled skills (`find-skills`, `self-improving-agent`, `tavily-search`, `brave-web-search`) are also enabled by default; if required API keys are missing, OpenClaw will surface configuration errors in runtime.  \nThe Skills page can display skills discovered from multiple OpenClaw sources (managed dir, workspace, and extra skill dirs), and now shows each skill's actual location so you can open the real folder directly.\n\nEnvironment variables for bundled search skills:\n- `BRAVE_SEARCH_API_KEY` for `brave-web-search`\n- `TAVILY_API_KEY` for `tavily-search` (OAuth may also be supported by upstream skill runtime)\n- `find-skills` and `self-improving-agent` do not require API keys\n\n### 🔐 Secure Provider Integration\nConnect to multiple AI providers (OpenAI, Anthropic, and more) with credentials stored securely in your system's native keychain. OpenAI supports both API key and browser OAuth (Codex subscription) sign-in.\nFor **Custom** providers used with OpenAI-compatible gateways, you can set a custom `User-Agent` in **Settings → AI Providers → Edit Provider** for compatibility-sensitive endpoints.\n\n### 🌙 Adaptive Theming\nLight mode, dark mode, or system-synchronized themes. ClawX adapts to your preferences automatically.\n\n### 🚀 Startup Launch Control\nIn **Settings → General**, you can enable **Launch at system startup** so ClawX starts automatically after login.\n\n---\n\n## Getting Started\n\n### System Requirements\n\n- **Operating System**: macOS 11+, Windows 10+, or Linux (Ubuntu 20.04+)\n- **Memory**: 4GB RAM minimum (8GB recommended)\n- **Storage**: 1GB available disk space\n\n### Installation\n\n#### Pre-built Releases (Recommended)\n\nDownload the latest release for your platform from the [Releases](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Freleases) page.\n\n#### Build from Source\n\n```bash\n# Clone the repository\ngit clone https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX.git\ncd ClawX\n\n# Initialize the project\npnpm run init\n\n# Start in development mode\npnpm dev\n```\n### First Launch\n\nWhen you launch ClawX for the first time, the **Setup Wizard** will guide you through:\n\n1. **Language & Region** – Configure your preferred locale\n2. **AI Provider** – Add providers with API keys or OAuth (for providers that support browser\u002Fdevice login)\n3. **Skill Bundles** – Select pre-configured skills for common use cases\n4. **Verification** – Test your configuration before entering the main interface\n\nThe wizard preselects your system language when it is supported, and falls back to English otherwise.\n\n> Note for Moonshot (Kimi): ClawX keeps Kimi web search enabled by default.  \n> When Moonshot is configured, ClawX also syncs Kimi web search to the China endpoint (`https:\u002F\u002Fapi.moonshot.cn\u002Fv1`) in OpenClaw config.\n\n### Proxy Settings\n\nClawX includes built-in proxy settings for environments where Electron, the OpenClaw Gateway, or channels such as Telegram need to reach the internet through a local proxy client.\n\nOpen **Settings → Gateway → Proxy** and configure:\n\n- **Proxy Server**: the default proxy for all requests\n- **Bypass Rules**: hosts that should connect directly, separated by semicolons, commas, or new lines\n- In **Developer Mode**, you can optionally override:\n  - **HTTP Proxy**\n  - **HTTPS Proxy**\n  - **ALL_PROXY \u002F SOCKS**\n\nRecommended local examples:\n\n```text\nProxy Server: http:\u002F\u002F127.0.0.1:7890\n```\nNotes:\n\n- A bare `host:port` value is treated as HTTP.\n- If advanced proxy fields are left empty, ClawX falls back to `Proxy Server`.\n- Saving proxy settings reapplies Electron networking immediately and restarts the Gateway automatically.\n- ClawX also syncs the proxy to OpenClaw's Telegram channel config when Telegram is enabled.\n- Gateway restarts preserve an existing Telegram channel proxy if ClawX proxy is currently disabled.\n- To explicitly clear Telegram channel proxy from OpenClaw config, save proxy settings with proxy disabled.\n- In **Settings → Advanced → Developer**, you can run **OpenClaw Doctor** to execute `openclaw doctor --json` and inspect the diagnostic output without leaving the app.\n- On packaged Windows builds, the bundled `openclaw` CLI\u002FTUI runs via the shipped `node.exe` entrypoint to keep terminal input behavior stable.\n\n---\n\n## Architecture\n\nClawX employs a **dual-process architecture** with a unified host API layer. The renderer talks to a single client abstraction, while Electron Main owns protocol selection and process lifecycle:\n\n```┌─────────────────────────────────────────────────────────────────┐\n│                        ClawX Desktop App                         │\n│                                                                  │\n│  ┌────────────────────────────────────────────────────────────┐  │\n│  │              Electron Main Process                          │  │\n│  │  • Window & application lifecycle management               │  │\n│  │  • Gateway process supervision                              │  │\n│  │  • System integration (tray, notifications, keychain)       │  │\n│  │  • Auto-update orchestration                                │  │\n│  └────────────────────────────────────────────────────────────┘  │\n│                              │                                    │\n│                              │ IPC (authoritative control plane)  │\n│                              ▼                                    │\n│  ┌────────────────────────────────────────────────────────────┐  │\n│  │              React Renderer Process                         │  │\n│  │  • Modern component-based UI (React 19)                     │  │\n│  │  • State management with Zustand                            │  │\n│  │  • Unified host-api\u002Fapi-client calls                        │  │\n│  │  • Rich Markdown rendering                                  │  │\n│  └────────────────────────────────────────────────────────────┘  │\n└──────────────────────────────┬──────────────────────────────────┘\n                               │\n                               │ Main-owned transport strategy\n                               │ (WS first, HTTP then IPC fallback)\n                               ▼\n┌─────────────────────────────────────────────────────────────────┐\n│                Host API & Main Process Proxies                  │\n│                                                                  │\n│  • hostapi:fetch (Main proxy, avoids CORS in dev\u002Fprod)          │\n│  • gateway:httpProxy (Renderer never calls Gateway HTTP direct)  │\n│  • Unified error mapping & retry\u002Fbackoff                         │\n└──────────────────────────────┬──────────────────────────────────┘\n                               │\n                               │ WS \u002F HTTP \u002F IPC fallback\n                               ▼\n┌─────────────────────────────────────────────────────────────────┐\n│                     OpenClaw Gateway                             │\n│                                                                  │\n│  • AI agent runtime and orchestration                           │\n│  • Message channel management                                    │\n│  • Skill\u002Fplugin execution environment                           │\n│  • Provider abstraction layer                                    │\n└─────────────────────────────────────────────────────────────────┘\n```\n### Design Principles\n\n- **Process Isolation**: The AI runtime operates in a separate process, ensuring UI responsiveness even during heavy computation\n- **Single Entry for Frontend Calls**: Renderer requests go through host-api\u002Fapi-client; protocol details are hidden behind a stable interface\n- **Main-Process Transport Ownership**: Electron Main controls WS\u002FHTTP usage and fallback to IPC for reliability\n- **Graceful Recovery**: Built-in reconnect, timeout, and backoff logic handles transient failures automatically\n- **Secure Storage**: API keys and sensitive data leverage the operating system's native secure storage mechanisms\n- **CORS-Safe by Design**: Local HTTP access is proxied by Main, preventing renderer-side CORS issues\n\n### Process Model & Gateway Troubleshooting\n\n- ClawX is an Electron app, so **one app instance normally appears as multiple OS processes** (main\u002Frenderer\u002Fzygote\u002Futility). This is expected.\n- Single-instance protection uses Electron's lock plus a local process-file lock fallback, preventing duplicate app launch in environments where desktop IPC\u002Fsession bus is unstable.\n- During rolling upgrades, mixed old\u002Fnew app versions can still have asymmetric protection behavior. For best reliability, upgrade all desktop clients to the same version.\n- The OpenClaw Gateway listener should still be **single-owner**: only one process should listen on `127.0.0.1:18789`.\n- To verify the active listener:\n  - macOS\u002FLinux: `lsof -nP -iTCP:18789 -sTCP:LISTEN`\n  - Windows (PowerShell): `Get-NetTCPConnection -LocalPort 18789 -State Listen`\n- Clicking the window close button (`X`) hides ClawX to tray; it does **not** fully quit the app. Use tray menu **Quit ClawX** for complete shutdown.\n\n---\n\n## Use Cases\n\n### 🤖 Personal AI Assistant\nConfigure a general-purpose AI agent that can answer questions, draft emails, summarize documents, and help with everyday tasks—all from a clean desktop interface.\n\n### 📊 Automated Monitoring\nSet up scheduled agents to monitor news feeds, track prices, or watch for specific events. Results are delivered to your preferred notification channel.\n\n### 💻 Developer Productivity\nIntegrate AI into your development workflow. Use agents to review code, generate documentation, or automate repetitive coding tasks.\n\n### 🔄 Workflow Automation\nChain multiple skills together to create sophisticated automation pipelines. Process data, transform content, and trigger actions—all orchestrated visually.\n\n---\n\n## Development\n\n### Prerequisites\n\n- **Node.js**: 22+ (LTS recommended)\n- **Package Manager**: pnpm 9+ (recommended) or npm\n\n### Project Structure\n\n```ClawX\u002F\n├── electron\u002F                 # Electron Main Process\n│   ├── api\u002F                 # Main-side API router and handlers\n│   │   └── routes\u002F          # RPC\u002FHTTP proxy route modules\n│   ├── services\u002F            # Provider, secrets and runtime services\n│   │   ├── providers\u002F       # Provider\u002Faccount model sync logic\n│   │   └── secrets\u002F         # OS keychain and secret storage\n│   ├── shared\u002F              # Shared provider schemas\u002Fconstants\n│   │   └── providers\u002F\n│   ├── main\u002F                # App entry, windows, IPC registration\n│   ├── gateway\u002F             # OpenClaw Gateway process manager\n│   ├── preload\u002F             # Secure IPC bridge\n│   └── utils\u002F               # Utilities (storage, auth, paths)\n├── src\u002F                      # React Renderer Process\n│   ├── lib\u002F                 # Unified frontend API + error model\n│   ├── stores\u002F              # Zustand stores (settings\u002Fchat\u002Fgateway)\n│   ├── components\u002F          # Reusable UI components\n│   ├── pages\u002F               # Setup\u002FDashboard\u002FChat\u002FChannels\u002FSkills\u002FCron\u002FSettings\n│   ├── i18n\u002F                # Localization resources\n│   └── types\u002F               # TypeScript type definitions\n├── tests\u002F\n│   └── unit\u002F                # Vitest unit\u002Fintegration-like tests\n├── resources\u002F                # Static assets (icons\u002Fimages)\n└── scripts\u002F                  # Build and utility scripts\n```\n### Available Commands\n\n```bash\n# Development\npnpm run init             # Install dependencies + download uv\npnpm dev                  # Start with hot reload (auto-prepares bundled skills if missing)\n\n# Quality\npnpm lint                 # Run ESLint\npnpm typecheck            # TypeScript validation\n\n# Testing\npnpm test                 # Run unit tests\npnpm run comms:replay     # Compute communication replay metrics\npnpm run comms:baseline   # Refresh communication baseline snapshot\npnpm run comms:compare    # Compare replay metrics against baseline thresholds\n\n# Build & Package\npnpm run build:vite       # Build frontend only\npnpm build                # Full production build (with packaging assets)\npnpm package              # Package for current platform (includes bundled preinstalled skills)\npnpm package:mac          # Package for macOS\npnpm package:win          # Package for Windows\npnpm package:linux        # Package for Linux\n```\n\n### Communication Regression Checks\n\nWhen a PR changes communication paths (gateway events, chat runtime send\u002Freceive flow, channel delivery, or transport fallback), run:\n\n```bash\npnpm run comms:replay\npnpm run comms:compare\n```\n\n`comms-regression` in CI enforces required scenarios and threshold checks.\n### Tech Stack\n\n| Layer | Technology |\n|-------|------------|\n| Runtime | Electron 40+ |\n| UI Framework | React 19 + TypeScript |\n| Styling | Tailwind CSS + shadcn\u002Fui |\n| State | Zustand |\n| Build | Vite + electron-builder |\n| Testing | Vitest + Playwright |\n| Animation | Framer Motion |\n| Icons | Lucide React |\n\n---\n\n## Contributing\n\nWe welcome contributions from the community! Whether it's bug fixes, new features, documentation improvements, or translations—every contribution helps make ClawX better.\n\n### How to Contribute\n\n1. **Fork** the repository\n2. **Create** a feature branch (`git checkout -b feature\u002Famazing-feature`)\n3. **Commit** your changes with clear messages\n4. **Push** to your branch\n5. **Open** a Pull Request\n\n### Guidelines\n\n- Follow the existing code style (ESLint + Prettier)\n- Write tests for new functionality\n- Update documentation as needed\n- Keep commits atomic and descriptive\n\n---\n\n## Acknowledgments\n\nClawX is built on the shoulders of excellent open-source projects:\n\n- [OpenClaw](https:\u002F\u002Fgithub.com\u002FOpenClaw) – The AI agent runtime\n- [Electron](https:\u002F\u002Fwww.electronjs.org\u002F) – Cross-platform desktop framework\n- [React](https:\u002F\u002Freact.dev\u002F) – UI component library\n- [shadcn\u002Fui](https:\u002F\u002Fui.shadcn.com\u002F) – Beautifully designed components\n- [Zustand](https:\u002F\u002Fgithub.com\u002Fpmndrs\u002Fzustand) – Lightweight state management\n\n---\n\n## Community\n\nJoin our community to connect with other users, get support, and share your experiences.\n\n| Enterprise WeChat | Feishu Group | Discord |\n| :---: | :---: | :---: |\n| \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_5641d55ffe1c.png\" width=\"150\" alt=\"WeChat QR Code\" \u002F> | \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_a9b585611ba5.png\" width=\"150\" alt=\"Feishu QR Code\" \u002F> | \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_4ef6cdc7f7cb.png\" width=\"150\" alt=\"Discord QR Code\" \u002F> |\n\n### ClawX Partner Program 🚀\n\nWe're launching the ClawX Partner Program and looking for partners who can help introduce ClawX to more clients, especially those with custom AI agent or automation needs.\n\nPartners help connect us with potential users and projects, while the ClawX team provides full technical support, customization, and integration.\n\nIf you work with clients interested in AI tools or automation, we'd love to collaborate.\n\nDM us or email [public@valuecell.ai](mailto:public@valuecell.ai) to learn more.\n\n---\n\n## Star History\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_3612ffede7d4.png\" alt=\"Star History Chart\" \u002F>\n\u003C\u002Fp>\n\n---\n\n## License\n\nClawX is released under the [MIT License](LICENSE). You're free to use, modify, and distribute this software.\n\n---\n\n\u003Cp align=\"center\">\n  \u003Csub>Built with ❤️ by the ValueCell Team\u003C\u002Fsub>\n\u003C\u002Fp>\n","\u003Cp align=\"center\">\n  \u003Cimg src=\"src\u002Fassets\u002Flogo.svg\" width=\"128\" height=\"128\" alt=\"ClawX Logo\" \u002F>\n\u003C\u002Fp>\n\n\u003Ch1 align=\"center\">ClawX\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>OpenClaw AI 智能体的桌面界面\u003C\u002Fstrong>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"#features\">功能特性\u003C\u002Fa> •\n  \u003Ca href=\"#why-clawx\">为何选择 ClawX\u003C\u002Fa> •\n  \u003Ca href=\"#getting-started\">快速开始\u003C\u002Fa> •\n  \u003Ca href=\"#architecture\">架构\u003C\u002Fa> •\n  \u003Ca href=\"#development\">开发\u003C\u002Fa> •\n  \u003Ca href=\"#contributing\">贡献\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fplatform-MacOS%20%7C%20Windows%20%7C%20Linux-blue\" alt=\"Platform\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Felectron-40+-47848F?logo=electron\" alt=\"Electron\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Freact-19-61DAFB?logo=react\" alt=\"React\" \u002F>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.com\u002Finvite\u002F84Kex3GGAh\" target=\"_blank\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fdiscord\u002F1399603591471435907?logo=discord&labelColor=%20%235462eb&logoColor=%20%23f5f5f5&color=%20%235462eb\" alt=\"chat on Discord\" \u002F>\n  \u003C\u002Fa>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fdownloads\u002FValueCell-ai\u002FClawX\u002Ftotal?color=%23027DEB\" alt=\"Downloads\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-green\" alt=\"License\" \u002F>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  English | \u003Ca href=\"README.zh-CN.md\">简体中文\u003C\u002Fa> | \u003Ca href=\"README.ja-JP.md\">日本語\u003C\u002Fa>\n\u003C\u002Fp>\n\n---\n\n## 概述\n\n**ClawX** 弥合了强大的 AI 智能体（AI Agents）与普通用户之间的鸿沟。基于 [OpenClaw](https:\u002F\u002Fgithub.com\u002FOpenClaw) 构建，它将命令行的 AI 编排（orchestration）转化为直观、美观的桌面体验——无需使用终端。\n\n无论你是自动化工作流、管理 AI 驱动的频道（channels），还是调度智能任务，ClawX 都为你提供了高效驾驭 AI 智能体所需的界面。\n\nClawX 预配置了业界最佳实践的模型提供商（model providers），并原生支持 Windows 及多语言设置。当然，你也可以通过 **设置 → 高级 → 开发者模式** 进行高级配置微调。\n\n---\n## 截图\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_56a89596f02d.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_48e6d99132e7.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_3e576ab04cac.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_0cbe90d9b324.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_e51903d71a7e.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_46cf0b0d3c0f.png\" style=\"width: 100%; height: auto;\">\n\u003C\u002Fp>\n\n---\n\n## 为何选择 ClawX\n\n构建 AI 智能体不应要求你精通命令行。ClawX 的设计理念很简单：**强大的技术值得拥有一个尊重你时间的界面。**\n\n| 挑战 | ClawX 解决方案 |\n|-----------|----------------|\n| 复杂的 CLI 设置 | 一键安装 + 引导式设置向导 |\n| 配置文件 | 可视化设置 + 实时验证 |\n| 进程管理 | 自动网关生命周期管理 |\n| 多个 AI 提供商 | 统一的提供商配置面板 |\n| 技能\u002F插件安装 | 内置技能市场与管理 |\n\n### 内置 OpenClaw\n\nClawX 直接基于官方 **OpenClaw** 核心构建。我们无需用户单独安装，而是将运行时（runtime）嵌入应用内部，提供“开箱即用”的无缝体验。\n\n我们致力于与上游 OpenClaw 项目保持严格同步，确保你始终能获得官方版本提供的最新功能、稳定性改进和生态系统兼容性。\n\n---\n\n## 功能特性\n\n### 🎯 零配置门槛\n从安装到首次与 AI 交互，全程通过直观的图形界面完成。无需终端命令、无需 YAML 文件、无需手动查找环境变量。\n\n### 💬 智能聊天界面\n通过现代化的聊天体验与 AI 智能体沟通。支持多会话上下文、消息历史、Markdown 富内容渲染，以及在主输入框中直接使用 `@agent` 路由实现多智能体交互。  \n当你使用 `@agent` 指向另一个智能体时，ClawX 会直接切换到该智能体自身的会话上下文中，而非通过默认智能体中转。智能体工作区默认相互隔离，更强的隔离性取决于 OpenClaw 的沙箱（sandbox）设置。  \n每个智能体还可覆盖自身的 `provider\u002Fmodel` 运行时设置；未覆盖的智能体将继续继承全局默认模型。\n\n### 📡 多频道管理\n同时配置和监控多个 AI 频道。每个频道独立运行，让你为不同任务运行专用智能体。  \n每个频道现在支持多个账号、按账号绑定智能体，并可直接在“频道”页面切换频道的默认账号。  \nClawX 现已内置腾讯官方个人微信频道插件，你可通过应用内二维码流程直接从“频道”页面绑定微信。\n\n### ⏰ 基于 Cron 的自动化\n调度 AI 任务自动运行。定义触发条件、设置执行间隔，让你的 AI 智能体全天候自动工作，无需人工干预。  \n“Cron” 页面现在允许你在任务表单中直接配置外部投递（external delivery），包含独立的发送账号和接收目标选择器。对于支持的频道，接收目标会自动从频道目录或已知会话历史中发现，因此你不再需要手动编辑 `jobs.json`。  \n已知限制：目前微信被有意排除在支持的 Cron 投递频道之外。当前的 `openclaw-weixin` 插件要求外发消息必须有活跃会话的 `contextToken`，因此该插件本身不支持 Cron 式的主动投递。\n\n### 🧩 可扩展技能系统（Extensible Skill System）\n通过预构建的技能扩展你的 AI 智能体。通过集成的技能面板浏览、安装和管理技能——无需使用包管理器。\n\nClawX 还预置了完整的文档处理技能（`pdf`、`xlsx`、`docx`、`pptx`），在启动时会自动将它们部署到托管技能目录（默认为 `~\u002F.openclaw\u002Fskills`），并在首次安装时默认启用。其他捆绑技能（`find-skills`、`self-improving-agent`、`tavily-search`、`brave-web-search`）也默认启用；如果缺少所需的 API 密钥，OpenClaw 会在运行时提示配置错误。\n\n技能页面现在可以显示从多个 OpenClaw 来源（托管目录、工作区和额外技能目录）发现的技能，并显示每个技能的实际位置，以便你可以直接打开真实文件夹。\n\n捆绑搜索技能所需的环境变量：\n- `BRAVE_SEARCH_API_KEY` 用于 `brave-web-search`\n- `TAVILY_API_KEY` 用于 `tavily-search`（上游技能运行时可能也支持 OAuth）\n- `find-skills` 和 `self-improving-agent` 不需要 API 密钥\n\n### 🔐 安全的提供商集成（Secure Provider Integration）\n安全地连接多个 AI 提供商（如 OpenAI、Anthropic 等），凭据存储在系统原生密钥链中。OpenAI 同时支持 API 密钥和浏览器 OAuth（Codex 订阅）登录方式。\n\n对于与 OpenAI 兼容网关一起使用的**自定义（Custom）**提供商，你可以在 **设置 → AI 提供商 → 编辑提供商** 中设置自定义 `User-Agent`，以兼容对 User-Agent 敏感的端点。\n\n### 🌙 自适应主题（Adaptive Theming）\n支持浅色模式、深色模式或跟随系统主题。ClawX 会自动适配你的偏好设置。\n\n### 🚀 启动控制（Startup Launch Control）\n在 **设置 → 常规** 中，你可以启用 **系统启动时自动启动**，使 ClawX 在登录后自动运行。\n\n---\n\n## 快速开始\n\n### 系统要求\n\n- **操作系统**：macOS 11+、Windows 10+ 或 Linux（Ubuntu 20.04+）\n- **内存**：最低 4GB RAM（推荐 8GB）\n- **存储空间**：至少 1GB 可用磁盘空间\n\n### 安装\n\n#### 预构建版本（推荐）\n\n从 [Releases](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Freleases) 页面下载适用于你平台的最新版本。\n\n#### 从源码构建\n\n```bash\n# 克隆仓库\ngit clone https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX.git\ncd ClawX\n\n# 初始化项目\npnpm run init\n\n# 以开发模式启动\npnpm dev\n```\n\n### 首次启动\n\n首次启动 ClawX 时，**设置向导（Setup Wizard）** 将引导你完成以下步骤：\n\n1. **语言与地区** – 配置首选区域设置\n2. **AI 提供商** – 添加提供商，可通过 API 密钥或 OAuth 登录（适用于支持浏览器\u002F设备登录的提供商）\n3. **技能包** – 为常见使用场景选择预配置的技能\n4. **验证** – 在进入主界面之前测试你的配置\n\n当系统语言受支持时，向导会自动预选该语言；否则回退为英语。\n\n> Moonshot（Kimi）注意事项：ClawX 默认启用 Kimi 网页搜索。  \n> 配置 Moonshot 后，ClawX 还会将 Kimi 网页搜索同步到中国区端点（`https:\u002F\u002Fapi.moonshot.cn\u002Fv1`）的 OpenClaw 配置中。\n\n### 代理设置\n\nClawX 内置代理设置，适用于 Electron、OpenClaw 网关或 Telegram 等通道需要通过本地代理客户端访问互联网的环境。\n\n打开 **设置 → 网关 → 代理** 并配置：\n\n- **代理服务器（Proxy Server）**：所有请求的默认代理\n- **绕过规则（Bypass Rules）**：应直连的主机，用分号、逗号或换行分隔\n- 在 **开发者模式（Developer Mode）** 下，你还可以选择性覆盖：\n  - **HTTP 代理**\n  - **HTTPS 代理**\n  - **ALL_PROXY \u002F SOCKS**\n\n推荐的本地示例：\n\n```text\nProxy Server: http:\u002F\u002F127.0.0.1:7890\n```\n\n注意事项：\n\n- 单独的 `host:port` 值会被视为 HTTP 代理。\n- 如果高级代理字段留空，ClawX 会回退使用 `Proxy Server`。\n- 保存代理设置会立即重新应用 Electron 网络配置，并自动重启网关。\n- 当启用 Telegram 时，ClawX 还会将代理同步到 OpenClaw 的 Telegram 通道配置中。\n- 网关重启时，如果当前 ClawX 代理已禁用，则保留现有的 Telegram 通道代理。\n- 要显式清除 OpenClaw 配置中的 Telegram 通道代理，请在禁用代理的情况下保存代理设置。\n- 在 **设置 → 高级 → 开发者** 中，你可以运行 **OpenClaw Doctor** 来执行 `openclaw doctor --json` 并在应用内查看诊断输出。\n- 在打包的 Windows 版本中，捆绑的 `openclaw` CLI\u002FTUI 通过附带的 `node.exe` 入口运行，以保持终端输入行为稳定。\n\n## 架构\n\nClawX 采用**双进程架构（dual-process architecture）**，并配备统一的宿主 API（host API）层。渲染器（renderer）与单一客户端抽象进行通信，而 Electron 主进程（Electron Main）负责协议选择和进程生命周期管理：\n\n```┌─────────────────────────────────────────────────────────────────┐\n│                        ClawX 桌面应用                            │\n│                                                                  │\n│  ┌────────────────────────────────────────────────────────────┐  │\n│  │              Electron 主进程                                │  │\n│  │  • 窗口与应用程序生命周期管理                               │  │\n│  │  • 网关进程监督                                              │  │\n│  │  • 系统集成（托盘、通知、钥匙串）                           │  │\n│  │  • 自动更新编排                                              │  │\n│  └────────────────────────────────────────────────────────────┘  │\n│                              │                                    │\n│                              │ IPC（权威控制平面）                │\n│                              ▼                                    │\n│  ┌────────────────────────────────────────────────────────────┐  │\n│  │              React 渲染进程                                 │  │\n│  │  • 基于现代组件的 UI（React 19）                            │  │\n│  │  • 使用 Zustand 进行状态管理                                │  │\n│  │  • 统一的 host-api\u002Fapi-client 调用                          │  │\n│  │  • 丰富的 Markdown 渲染                                     │  │\n│  └────────────────────────────────────────────────────────────┘  │\n└──────────────────────────────┬──────────────────────────────────┘\n                               │\n                               │ 由主进程拥有的传输策略\n                               │ （优先 WebSocket，其次 HTTP，最后回退到 IPC）\n                               ▼\n┌─────────────────────────────────────────────────────────────────┐\n│                宿主 API 与主进程代理                             │\n│                                                                  │\n│  • hostapi:fetch（主进程代理，避免开发\u002F生产环境中的 CORS 问题） │  │\n│  • gateway:httpProxy（渲染器从不直接调用网关 HTTP）             │  │\n│  • 统一的错误映射与重试\u002F退避机制                                │  │\n└──────────────────────────────┬──────────────────────────────────┘\n                               │\n                               │ WebSocket \u002F HTTP \u002F IPC 回退\n                               ▼\n┌─────────────────────────────────────────────────────────────────┐\n│                     OpenClaw 网关（Gateway）                    │\n│                                                                  │\n│  • AI 代理运行时与编排                                           │  │\n│  • 消息通道管理                                                  │  │\n│  • 技能\u002F插件执行环境                                             │  │\n│  • 提供商（Provider）抽象层                                      │  │\n└─────────────────────────────────────────────────────────────────┘\n```\n\n### 设计原则\n\n- **进程隔离（Process Isolation）**：AI 运行时在独立进程中运行，即使在高负载计算期间也能确保 UI 的响应性  \n- **前端调用单一入口（Single Entry for Frontend Calls）**：渲染器请求通过 host-api\u002Fapi-client 发起；协议细节被隐藏在稳定的接口之后  \n- **主进程拥有传输控制权（Main-Process Transport Ownership）**：Electron 主进程控制 WebSocket\u002FHTTP 的使用，并在必要时回退到 IPC 以提高可靠性  \n- **优雅恢复（Graceful Recovery）**：内置自动重连、超时和退避逻辑，可自动处理瞬时故障  \n- **安全存储（Secure Storage）**：API 密钥和敏感数据利用操作系统原生的安全存储机制  \n- **设计上规避 CORS（CORS-Safe by Design）**：本地 HTTP 访问由主进程代理，防止渲染器端出现 CORS 问题  \n\n### 进程模型与网关故障排查\n\n- ClawX 是一个 Electron 应用，因此**单个应用实例通常表现为多个操作系统进程**（主进程、渲染进程、zygote、utility 等），这是正常现象。  \n- 单实例保护机制结合了 Electron 的锁机制和本地进程文件锁作为后备方案，可在桌面 IPC\u002F会话总线不稳定的环境中防止重复启动应用。  \n- 在滚动升级过程中，新旧版本混合可能导致非对称的保护行为。为获得最佳可靠性，请将所有桌面客户端升级至相同版本。  \n- OpenClaw 网关监听器应保持**单一所有者（single-owner）**：仅允许一个进程监听 `127.0.0.1:18789`。  \n- 验证当前活跃监听器的方法：\n  - macOS\u002FLinux: `lsof -nP -iTCP:18789 -sTCP:LISTEN`\n  - Windows (PowerShell): `Get-NetTCPConnection -LocalPort 18789 -State Listen`\n- 点击窗口关闭按钮（`X`）会将 ClawX 隐藏到系统托盘；**不会**完全退出应用。如需彻底关闭，请使用托盘菜单中的 **Quit ClawX**。\n\n---\n\n## 使用场景\n\n### 🤖 个人 AI 助手  \n配置一个通用型 AI 代理，可用于回答问题、撰写邮件、总结文档以及协助处理日常任务——全部通过简洁的桌面界面完成。\n\n### 📊 自动化监控  \n设置定时代理，用于监控新闻源、追踪价格或监视特定事件。结果将发送至您偏好的通知渠道。\n\n### 💻 开发者效率提升  \n将 AI 集成到您的开发工作流中。使用代理审查代码、生成文档或自动化重复性编码任务。\n\n### 🔄 工作流自动化  \n将多个技能串联起来，创建复杂的工作流自动化管道。处理数据、转换内容并触发操作——全部通过可视化方式编排。\n\n---\n\n## 开发\n\n### 先决条件\n\n- **Node.js**: 22+（推荐使用 LTS 版本）  \n- **包管理器**: pnpm 9+（推荐）或 npm  \n\n### 项目结构\n\n```ClawX\u002F\n├── electron\u002F                 # Electron 主进程\n│   ├── api\u002F                 # 主进程侧 API 路由器和处理器\n│   │   └── routes\u002F          # RPC\u002FHTTP 代理路由模块\n│   ├── services\u002F            # 提供商、密钥和运行时服务\n│   │   ├── providers\u002F       # 提供商\u002F账户模型同步逻辑\n│   │   └── secrets\u002F         # 操作系统钥匙串与密钥存储\n│   ├── shared\u002F              # 共享的提供商 Schema\u002F常量\n│   │   └── providers\u002F\n│   ├── main\u002F                # 应用入口、窗口管理、IPC 注册\n│   ├── gateway\u002F             # OpenClaw 网关进程管理器\n│   ├── preload\u002F             # 安全 IPC 桥接\n│   └── utils\u002F               # 工具函数（存储、认证、路径等）\n├── src\u002F                      # React 渲染进程\n│   ├── lib\u002F                 # 统一的前端 API + 错误模型\n│   ├── stores\u002F              # Zustand 状态存储（设置\u002F聊天\u002F网关）\n│   ├── components\u002F          # 可复用 UI 组件\n│   ├── pages\u002F               # 设置\u002F仪表盘\u002F聊天\u002F频道\u002F技能\u002F定时任务\u002F设置页面\n│   ├── i18n\u002F                # 本地化资源\n│   └── types\u002F               # TypeScript 类型定义\n├── tests\u002F\n│   └── unit\u002F                # Vitest 单元测试\u002F类集成测试\n├── resources\u002F                # 静态资源（图标\u002F图片）\n└── scripts\u002F                  # 构建与工具脚本\n```\n\n### 可用命令\n\n```bash\n# 开发\npnpm run init             # 安装依赖 + 下载 uv\npnpm dev                  # 启动开发服务器（带热重载，若缺少打包的技能会自动准备）\n\n# 代码质量\npnpm lint                 # 运行 ESLint\npnpm typecheck            # TypeScript 类型检查\n\n# 测试\npnpm test                 # 运行单元测试\npnpm run comms:replay     # 计算通信回放指标\npnpm run comms:baseline   # 刷新通信基准快照\npnpm run comms:compare    # 将回放指标与基准阈值进行比较\n\n# 构建与打包\npnpm run build:vite       # 仅构建前端\npnpm build                # 完整生产构建（包含打包资源）\npnpm package              # 为当前平台打包（包含预装的打包技能）\npnpm package:mac          # 为 macOS 打包\npnpm package:win          # 为 Windows 打包\npnpm package:linux        # 为 Linux 打包\n```\n\n### 通信回归检查\n\n当 PR 修改了通信路径（如网关事件、聊天运行时的发送\u002F接收流程、频道投递或传输降级机制）时，请运行：\n\n```bash\npnpm run comms:replay\npnpm run comms:compare\n```\n\nCI 中的 `comms-regression` 会强制执行所需场景和阈值检查。\n\n### 技术栈\n\n| 层级 | 技术 |\n|------|------|\n| 运行时（Runtime） | Electron 40+ |\n| UI 框架 | React 19 + TypeScript |\n| 样式 | Tailwind CSS + shadcn\u002Fui |\n| 状态管理 | Zustand |\n| 构建工具 | Vite + electron-builder |\n| 测试 | Vitest + Playwright |\n| 动画 | Framer Motion |\n| 图标 | Lucide React |\n\n---\n\n## 贡献指南\n\n我们欢迎社区贡献！无论是修复 bug、新增功能、改进文档还是翻译——每一份贡献都能让 ClawX 变得更好。\n\n### 如何贡献\n\n1. **Fork** 本仓库  \n2. **创建** 功能分支（`git checkout -b feature\u002Famazing-feature`）  \n3. **提交** 清晰明了的变更信息  \n4. **推送** 到你的分支  \n5. **发起** Pull Request  \n\n### 贡献规范\n\n- 遵循现有代码风格（ESLint + Prettier）  \n- 为新功能编写测试  \n- 按需更新文档  \n- 保持提交原子性且描述清晰  \n\n---\n\n## 致谢\n\nClawX 建立在众多优秀的开源项目之上：\n\n- [OpenClaw](https:\u002F\u002Fgithub.com\u002FOpenClaw) – AI 智能体运行时  \n- [Electron](https:\u002F\u002Fwww.electronjs.org\u002F) – 跨平台桌面应用框架  \n- [React](https:\u002F\u002Freact.dev\u002F) – UI 组件库  \n- [shadcn\u002Fui](https:\u002F\u002Fui.shadcn.com\u002F) – 设计精美的组件库  \n- [Zustand](https:\u002F\u002Fgithub.com\u002Fpmndrs\u002Fzustand) – 轻量级状态管理库  \n\n---\n\n## 社区\n\n加入我们的社区，与其他用户交流、获取支持并分享使用体验。\n\n| 企业微信 | 飞书群 | Discord |\n| :---: | :---: | :---: |\n| \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_5641d55ffe1c.png\" width=\"150\" alt=\"WeChat QR Code\" \u002F> | \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_a9b585611ba5.png\" width=\"150\" alt=\"Feishu QR Code\" \u002F> | \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_4ef6cdc7f7cb.png\" width=\"150\" alt=\"Discord QR Code\" \u002F> |\n\n### ClawX 合作伙伴计划 🚀\n\n我们即将启动 **ClawX 合作伙伴计划**，正在寻找能够帮助我们将 ClawX 推荐给更多客户（尤其是有定制化 AI 智能体或自动化需求的客户）的合作伙伴。\n\n合作伙伴负责帮我们对接潜在用户和项目，而 ClawX 团队将提供全面的技术支持、定制化服务和集成协助。\n\n如果你的客户对 AI 工具或自动化感兴趣，我们非常期待与你合作！\n\n请私信我们或发送邮件至 [public@valuecell.ai](mailto:public@valuecell.ai) 了解详情。\n\n---\n\n## Star 历史\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_readme_3612ffede7d4.png\" alt=\"Star History Chart\" \u002F>\n\u003C\u002Fp>\n\n---\n\n## 许可证\n\nClawX 采用 [MIT 许可证](LICENSE) 发布。你可以自由地使用、修改和分发本软件。\n\n---\n\n\u003Cp align=\"center\">\n  \u003Csub>由 ValueCell 团队 ❤️ 打造\u003C\u002Fsub>\n\u003C\u002Fp>","# ClawX 快速上手指南\n\n## 环境准备\n\n### 系统要求\n- **操作系统**：macOS 11+ \u002F Windows 10+ \u002F Linux（推荐 Ubuntu 20.04+）\n- **内存**：最低 4GB RAM（建议 8GB）\n- **存储空间**：至少 1GB 可用磁盘空间\n\n### 前置依赖（仅源码构建需要）\n- Node.js ≥ 18\n- pnpm 包管理器\n- Git\n\n> 💡 国内用户建议配置 npm \u002F pnpm 镜像源加速安装：\n> ```bash\n> pnpm config set registry https:\u002F\u002Fregistry.npmmirror.com\n> ```\n\n## 安装步骤\n\n### 方式一：下载预编译版本（推荐）\n前往 [GitHub Releases 页面](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Freleases)，根据你的操作系统下载对应安装包，直接运行即可。\n\n### 方式二：从源码构建\n```bash\n# 克隆仓库\ngit clone https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX.git\ncd ClawX\n\n# 初始化项目（自动安装依赖）\npnpm run init\n\n# 启动开发模式\npnpm dev\n```\n\n> 📌 国内网络环境下，若 `pnpm install` 缓慢，可先设置镜像源再执行 `pnpm run init`。\n\n## 基本使用\n\n1. **首次启动**  \n   安装完成后首次运行 ClawX，将自动弹出**设置向导**，按提示完成以下配置：\n   - 选择语言（支持简体中文）\n   - 添加 AI 提供商（如 OpenAI、Moonshot\u002FKimi 等），输入 API Key 或通过 OAuth 登录\n   - 选择预置技能包（如文档处理、网页搜索等）\n   - 验证配置并进入主界面\n\n2. **与 AI 交互**  \n   在主聊天界面直接输入消息即可与默认 AI 代理对话。  \n   若需调用其他代理，使用 `@代理名` 语法（例如 `@search`）即可切换上下文。\n\n3. **启用常用功能**\n   - **多渠道接入**：在「Channels」页面添加微信、Telegram 等通道（内置腾讯官方微信插件）\n   - **定时任务**：在「Cron」页面创建自动化任务（注意：微信暂不支持主动推送）\n   - **管理技能**：在「Skills」页面查看或启用预装技能（如 PDF 解析、Brave\u002FTavily 搜索）\n\n4. **代理设置（国内用户常用）**  \n   如需通过代理访问外网，在 **设置 → Gateway → Proxy** 中填写：\n   ```\n   Proxy Server: http:\u002F\u002F127.0.0.1:7890\n   ```\n   保存后会自动重启网关并生效。\n\n> ✅ 默认已启用 Kimi（Moonshot）的网页搜索功能，并自动配置为中国区端点 `https:\u002F\u002Fapi.moonshot.cn\u002Fv1`。","某跨境电商运营团队需要每天自动分析海外社交媒体评论情感，并将负面反馈汇总到内部系统，但他们缺乏专职开发人员，只能依赖开源 AI 工具链手动搭建流程。\n\n### 没有 ClawX 时\n- 必须在终端逐行输入命令安装 OpenClaw 及其依赖，Windows 环境下常因路径或权限问题失败  \n- 配置 AI 模型提供商（如 Ollama、OpenRouter）需手动编辑 YAML 文件，拼写错误导致服务无法启动  \n- 定时任务（如每日抓取评论）需编写 cron 表达式并通过 CLI 注册，调试困难且无可视化日志  \n- 添加“情感分析”技能插件需查找 GitHub 仓库、手动下载并放入指定目录，版本兼容性难以保证  \n- 多人协作时，配置文件需通过邮件或网盘共享，容易出现环境不一致问题  \n\n### 使用 ClawX 后\n- 通过图形化安装向导一键部署完整运行环境，5 分钟内完成从零到首次运行  \n- 在“模型”面板中直接选择并测试不同 AI 提供商，参数自动校验，无需接触配置文件  \n- 在“定时任务”界面用日历组件设置执行频率，实时查看任务历史与输出结果  \n- 通过内置“技能市场”一键安装官方认证的情感分析插件，自动处理依赖与更新  \n- 团队成员可导出\u002F导入整个项目配置包（含模型、技能、频道设置），确保环境完全一致  \n\nClawX 将原本面向开发者的技术栈转化为运营人员也能高效使用的桌面工作流，真正让 AI 自动化落地于日常业务。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FValueCell-ai_ClawX_56a89596.png","ValueCell-ai","ValueCell.ai","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FValueCell-ai_87bfbce1.png","",null,"public@valuecell.ai","valuecell","valuecell.ai","https:\u002F\u002Fgithub.com\u002FValueCell-ai",[85,89,93,96,100,104,108,111,114],{"name":86,"color":87,"percentage":88},"TypeScript","#3178c6",94,{"name":90,"color":91,"percentage":92},"JavaScript","#f1e05a",4.9,{"name":94,"color":79,"percentage":95},"NSIS",0.4,{"name":97,"color":98,"percentage":99},"Shell","#89e051",0.3,{"name":101,"color":102,"percentage":103},"PowerShell","#012456",0.2,{"name":105,"color":106,"percentage":107},"CSS","#663399",0.1,{"name":109,"color":110,"percentage":107},"Python","#3572A5",{"name":112,"color":113,"percentage":107},"Batchfile","#C1F12E",{"name":115,"color":116,"percentage":117},"HTML","#e34c26",0,6146,860,"2026-04-05T11:00:29","MIT","Linux, macOS, Windows","未说明","4GB minimum (8GB recommended)",{"notes":126,"python":123,"dependencies":127},"应用基于 Electron 和 React 构建，内嵌 OpenClaw 运行时，无需单独安装；支持系统级代理设置和多语言界面；首次启动会引导用户完成 AI 提供商配置和技能包选择；部分功能（如微信渠道的定时任务）存在限制。",[128,129,130,131],"Electron 40+","React 19","OpenClaw","pnpm",[14,13,15],[134,135,136,137,138,139,140,141],"agent","clawdbot","moltbot","openclaw","skill","agentic-ai","agents","ai",6,"2026-03-27T02:49:30.150509","2026-04-06T06:45:54.350338",[146,151,156,161,166,171,176,181],{"id":147,"question_zh":148,"answer_zh":149,"source_url":150},892,"Telegram 代理设置在重启 ClawX 后丢失怎么办？","该问题已在 v0.2.7 及更高版本中修复。升级到 v0.2.9 或更新版本后，Telegram 代理设置（如 http:\u002F\u002F127.0.0.1:XXXX）将被正确保留，不会因网关重启而丢失。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F501",{"id":152,"question_zh":153,"answer_zh":154,"source_url":155},893,"Windows 上开发者命令行无响应或命令被清除怎么办？","在 Windows 系统上，调用 openclaw 命令时可能会被直接清除导致无响应。建议优先使用最新版本（v0.2.7+），并避免重复启动多个 ClawX 实例。若仍存在问题，可尝试通过开发者模式手动执行命令。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F433",{"id":157,"question_zh":158,"answer_zh":159,"source_url":160},894,"ClawX 频繁卡死或网关无响应是什么原因？","常见原因是系统中存在多个 ClawX 进程同时运行，它们会竞争监听同一网关端口（默认 18789），导致端口冲突和网关崩溃。请检查任务管理器，结束所有 ClawX 进程后重新启动单个实例。新版本（如 v0.2.6-alpha.2 之后）已改善此问题。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F561",{"id":162,"question_zh":163,"answer_zh":164,"source_url":165},895,"安装 ClawX 后系统环境变量 PATH 被清空了怎么办？","该问题已在新版中修复。请确保使用 v0.2.1 之后的版本（推荐 v0.2.7+）。如果已发生 PATH 被清空，请从系统备份或默认配置恢复 PATH，并重新安装最新版 ClawX。维护者确认在干净的 Windows 系统上安装不会再覆盖现有 PATH。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F440",{"id":167,"question_zh":168,"answer_zh":169,"source_url":170},896,"安装后显示“网关错误”如何解决？","可尝试清空或重置配置文件 openclaw.json（注意先备份），然后填入以下最小配置：\n{\n  \"browser\": {},\n  \"gateway\": {\n    \"port\": 18789\n  }\n}\n保存后重启 ClawX，程序会自动补全其他配置项。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F96",{"id":172,"question_zh":173,"answer_zh":174,"source_url":175},897,"Windows 安装时提示“Python installation failed with code 1”怎么办？","建议选择“仅为我安装”（即安装到当前用户目录，如 C:\\Users\\\u003C用户名>\\AppData\\Local\\Programs\\ClawX），而非为所有用户安装。即使跳过 Python 安装，ClawX 核心功能仍可运行，但部分插件可能受限。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F49",{"id":177,"question_zh":178,"answer_zh":179,"source_url":180},898,"自定义模型时报错“API rate limit reached”如何处理？","该问题已在 v0.1.17 版本中修复。建议升级到最新版。此外，可尝试更换为兼容的模型（如 Claude），并确保 API 密钥和速率限制配置正确。参考社区方案：使用支持的自定义模型接口替代默认配置。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F145",{"id":182,"question_zh":183,"answer_zh":184,"source_url":185},899,"ClawX 网关经常无故停止运作怎么办？","此问题在 v0.3.0 版本中已得到改善。建议升级到 v0.3 或更高版本。同时确保没有多个 ClawX 进程同时运行，并检查是否因浏览器技能访问网页导致阻塞。更新后多数用户反馈网关稳定性显著提升。","https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues\u002F616",[187,192,197,202,207,212,217,222,227,232,237,242,247,252,257,262,267,272,277,282],{"id":188,"version":189,"summary_zh":190,"released_at":191},100480,"v0.3.6","## 🚀 ClawX v0.3.6\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(gateway): terminate owned process before retry to prevent port conflict on Windows by @shuang5725 in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F724\n* fix(gateway): refine process termination and wait logic on retry by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F750\n* Revert \"fix(gateway): refine process termination and wait logic on retry\" by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F754\n* Revert \"fix(gateway): terminate owned process before retry to prevent port conflict on Windows\" by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F755\n* fix(gateway): make heartbeat observability-only to prevent false cascade restarts by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F762\n* v0.3.6 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F763\n\n## New Contributors\n* @shuang5725 made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F724\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.5...v0.3.6","2026-04-03T07:48:55",{"id":193,"version":194,"summary_zh":195,"released_at":196},100481,"v0.3.5","## 🚀 ClawX v0.3.5\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* [codex] Preserve stable snapshots and stabilize Electron e2e by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F734\n* Upgrade openclaw to 4.1 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F742\n* fix: Host API port conflict crashing startup on Windows by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F743\n* fix(cron): replace hardcoded channel name in no-delivery-targets message by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F744\n* fix: sync Ollama provider config to gateway runtime (fixes #448) by @octo-patch in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F745\n* fix: properly sync Ollama provider config to gateway runtime by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F747\n* v0.3.5 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F748\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.4...v0.3.5","2026-04-02T07:34:35",{"id":198,"version":199,"summary_zh":200,"released_at":201},100482,"v0.3.4","## 🚀 ClawX v0.3.4\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* Feat\u002Fupgrade openclaw by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F729\n* fix(gateway): prevent default built-in plugins from being disabled by explicit allowlists by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F737\n* v0.3.4 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F739\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.3...v0.3.4","2026-04-01T10:26:43",{"id":203,"version":204,"summary_zh":205,"released_at":206},100483,"v0.3.3","## 🚀 ClawX v0.3.3\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* [codex] fix auth-backed provider discovery by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F690\n* fix: clean up deleted provider state correctly by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F696\n* fix: avoid systemd-supervised gateway retry loop in owned launcher by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F700\n* add electron e2e harness and regression coverage by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F697\n* Fix token usage handling and developer proxy save UX by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F704\n* fix(installer): resolve upgrade hang caused by locked files in installation directory by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F711\n* Upgrade third party plugin by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F715\n* fix(chat): hide internal system messages from webchat by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F710\n* fix gateway restart handshake timeout by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F689\n* fix(installer): preserve user .openclaw directory during uninstall by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F717\n* fix(gateway): resolve startup hang caused by qqbot plugin manifest ID mismatch & stderr flood by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F718\n* fix(gateway): restore WebSocket heartbeat ping on Windows by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F722\n* v0.3.3 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F723\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.2...v0.3.3","2026-03-31T03:36:18",{"id":208,"version":209,"summary_zh":210,"released_at":211},100484,"v0.3.2","## 🚀 ClawX v0.3.2\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(electron): work around Node.js cpSync Unicode crash on Windows by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F686\n* fix: force-clean stale instance lock when Electron lock guarantees exclusivity by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F685\n* fix: ensure custom provider deletion targets correct runtime key by making key generation idempotent by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F691\n* v0.3.2 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F693\n* fix(gateway): skip heartbeat ping on Windows to prevent reconnect cascade by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F694\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.1...v0.3.2","2026-03-27T11:16:55",{"id":213,"version":214,"summary_zh":215,"released_at":216},100485,"v0.3.1","## 🚀 ClawX v0.3.1\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* feat: unify cron delivery account and target selection by @xinzezhu in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F642\n* feat(agent-model): add per-agent model override with default-reset UX and runtime sync by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F651\n* fix(installer): enhance installation logging and user feedback during setup by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F661\n* fix: prevent config overwrite, session history race, and streaming message loss by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F663\n* fix(security): mitigate GHSA-9gf9-7xcc-xcq9 & GHSA-vf6c-fgmq-xm78 + bug fixes by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F667\n* upgrade openclaw to 3.23 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F652\n* feat: upgrade MiniMax default model from M2.5 to M2.7 by @octo-patch in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F671\n* fix: add @sinclair\u002Ftypebox dep and fix delivery button text truncation by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F673\n* v0.3.1 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F674\n* fix: update MiniMax model display to M2.7 on OAuth re-login by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F676\n* chore: free disk space in macOS CI by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F677\n* chore: fix mac ci by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F678\n\n## New Contributors\n* @octo-patch made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F671\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.3.0...v0.3.1","2026-03-26T12:33:03",{"id":218,"version":219,"summary_zh":220,"released_at":221},100486,"v0.3.0","## 🚀 ClawX v0.3.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(providers): model list empty in settings panel (#581) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* fix: sanitize stale nested plugin paths in plugins.load.paths by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F608\n* fix: preserve telegram proxy on gateway restart after doctor by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F546\n* fix: add fsPath prefix for Windows Unicode paths in config-sync plugin copy by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F612\n* feat(ark): add Code Plan mode preset and guided setup by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F617\n* upgrade wecom plugin version to 2026.3.20 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F619\n* feat(channel): support weichat channel by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F620\n* chore(telemetry): stop sending gateway reconnect payloads to PostHog by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F623\n* feat(release): 0.2.9 by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F629\n* changed feishu group code by @arianachennn-design in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F630\n* fix: persist theme setting to main process store by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F628\n* Fix discord channel native by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F634\n* fix(provider): preserve custom headers and add custom-provider User-Agent setting by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F635\n* fix: clean up WhatsApp credentials on QR cancel to prevent ghost channel by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F637\n* fix: persist provider display state across restarts (fixes #624) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F633\n* chore: normalize structure and split ipc handlers by @GASOT-GIT in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F590\n* feat(agents): add option to inherit main agent workspace when creating new agent by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F639\n* Fix provider display by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F641\n* fix: use openclaw.json as single source of truth for provider list by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F649\n* v0.3.0 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F650\n\n## New Contributors\n* @kagura-agent made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* @GASOT-GIT made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F590\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.8...v0.3.0","2026-03-24T09:16:51",{"id":223,"version":224,"summary_zh":225,"released_at":226},100487,"v0.3.0-alpha.0","## 🚀 ClawX v0.3.0-alpha.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(providers): model list empty in settings panel (#581) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* fix: sanitize stale nested plugin paths in plugins.load.paths by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F608\n* fix: preserve telegram proxy on gateway restart after doctor by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F546\n* fix: add fsPath prefix for Windows Unicode paths in config-sync plugin copy by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F612\n* feat(ark): add Code Plan mode preset and guided setup by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F617\n* upgrade wecom plugin version to 2026.3.20 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F619\n* feat(channel): support weichat channel by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F620\n* chore(telemetry): stop sending gateway reconnect payloads to PostHog by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F623\n* feat(release): 0.2.9 by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F629\n* changed feishu group code by @arianachennn-design in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F630\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.8...v0.3.0-alpha.0","2026-03-23T06:26:50",{"id":228,"version":229,"summary_zh":230,"released_at":231},100488,"v0.2.9","## 🚀 ClawX v0.2.9\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(providers): model list empty in settings panel (#581) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* fix: sanitize stale nested plugin paths in plugins.load.paths by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F608\n* fix: preserve telegram proxy on gateway restart after doctor by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F546\n* fix: add fsPath prefix for Windows Unicode paths in config-sync plugin copy by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F612\n* feat(ark): add Code Plan mode preset and guided setup by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F617\n* upgrade wecom plugin version to 2026.3.20 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F619\n* feat(channel): support weichat channel by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F620\n* chore(telemetry): stop sending gateway reconnect payloads to PostHog by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F623\n\n## New Contributors\n* @kagura-agent made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.8...v0.2.9","2026-03-23T03:16:22",{"id":233,"version":234,"summary_zh":235,"released_at":236},100489,"v0.2.9-beta.1","## 🚀 ClawX v0.2.9-beta.1\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(providers): model list empty in settings panel (#581) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* fix: sanitize stale nested plugin paths in plugins.load.paths by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F608\n* fix: preserve telegram proxy on gateway restart after doctor by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F546\n* fix: add fsPath prefix for Windows Unicode paths in config-sync plugin copy by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F612\n* feat(ark): add Code Plan mode preset and guided setup by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F617\n* upgrade wecom plugin version to 2026.3.20 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F619\n* feat(channel): support weichat channel by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F620\n\n## New Contributors\n* @kagura-agent made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.8...v0.2.9-beta.1","2026-03-22T10:32:33",{"id":238,"version":239,"summary_zh":240,"released_at":241},100490,"v0.2.9-alpha.0","## 🚀 ClawX v0.2.9-alpha.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(providers): model list empty in settings panel (#581) by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n* fix: sanitize stale nested plugin paths in plugins.load.paths by @kagura-agent in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F608\n* fix: preserve telegram proxy on gateway restart after doctor by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F546\n* fix: add fsPath prefix for Windows Unicode paths in config-sync plugin copy by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F612\n\n## New Contributors\n* @kagura-agent made their first contribution in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F591\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.8...v0.2.9-alpha.0","2026-03-22T09:14:18",{"id":243,"version":244,"summary_zh":245,"released_at":246},100491,"v0.2.8","## 🚀 ClawX v0.2.8\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* chore: upgrade feishu plugin version to 3.17 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F596\n* chore: upgrade wecom plugin  to 1.0.13 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F598\n* chore: upgrade dingding plugin to 3.3.0 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F599\n* fix(processes): fix multiple clawx processes running concurently by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F589\n* fix(build): prevent node download script from deleting uv.exe on Windows by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F600\n* v0.2.8 by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F601\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.7...v0.2.8","2026-03-20T11:22:44",{"id":248,"version":249,"summary_zh":250,"released_at":251},100492,"v0.2.7","## 🚀 ClawX v0.2.7\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(linux): Can't change Chinese IMEs on Debian by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F582\n* fix(models): use useReducer for token usage fetch state to fix lint errors and reduce loading time by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F586\n* fix(windows): Gateway process install extension failed by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F587\n* fix(windows): run bundled openclaw CLI\u002FTUI via node.exe to restore terminal input by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F571\n* fix(gateway): harden heartbeat timeout recovery to avoid reconnect flapping by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F588\n* fix gateway restart by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F593\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5...v0.2.7","2026-03-20T03:09:30",{"id":253,"version":254,"summary_zh":255,"released_at":256},100493,"v0.2.7-beta.0","## 🚀 ClawX v0.2.7-beta.0\r\n\r\nClawX - Graphical AI Assistant based on OpenClaw\r\n\r\n### 📦 Downloads\r\n\r\nPlease select the appropriate installer for your operating system and architecture:\r\n\r\n#### macOS\r\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\r\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\r\n\r\n#### Windows\r\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\r\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\r\n\r\n#### Linux\r\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\r\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\r\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\r\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\r\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\r\n\r\n### 📝 Release Notes\r\n\r\nSee the auto-generated release notes below for detailed changes.\r\n\r\n### ⚠️ Installation Notes\r\n\r\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\r\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\r\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\r\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\r\n\r\n---\r\n\r\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\r\n\r\n\r\n## What's Changed\r\n* fix(gateway): harden heartbeat timeout recovery to avoid reconnect flapping by @ashione in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F588\r\n* fix gateway restart by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F593\r\n\r\n\r\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.7-alpha.0...v0.2.7-beta.0","2026-03-19T11:02:51",{"id":258,"version":259,"summary_zh":260,"released_at":261},100494,"v0.2.7-alpha.0","## 🚀 ClawX v0.2.7-alpha.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(windows): run bundled openclaw CLI\u002FTUI via node.exe to restore terminal input by @vcfgv in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F571\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.6-beta.0...v0.2.7-alpha.0","2026-03-19T06:05:25",{"id":263,"version":264,"summary_zh":265,"released_at":266},100495,"v0.2.6-beta.0","## 🚀 ClawX v0.2.6-beta.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(linux): Can't change Chinese IMEs on Debian by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F582\n* fix(models): use useReducer for token usage fetch state to fix lint errors and reduce loading time by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F586\n* fix(windows): Gateway process install extension failed by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F587\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5-beta.4...v0.2.6-beta.0","2026-03-19T05:50:05",{"id":268,"version":269,"summary_zh":270,"released_at":271},100496,"v0.2.6-alpha.2","## 🚀 ClawX v0.2.6-alpha.2\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(linux): Can't change Chinese IMEs on Debian by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F582\n* fix(models): use useReducer for token usage fetch state to fix lint errors and reduce loading time by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F586\n* fix(windows): Gateway process install extension failed by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F587\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5-beta.4...v0.2.6-alpha.2","2026-03-19T05:44:56",{"id":273,"version":274,"summary_zh":275,"released_at":276},100497,"v0.2.6-alpha.1","## 🚀 ClawX v0.2.6-alpha.1\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix(linux): Can't change Chinese IMEs on Debian by @hazeone in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F582\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5-beta.4...v0.2.6-alpha.1","2026-03-19T03:31:18",{"id":278,"version":279,"summary_zh":280,"released_at":281},100498,"v0.2.6-alpha.0","## 🚀 ClawX v0.2.6-alpha.0\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5-beta.4...v0.2.6-alpha.0","2026-03-18T12:05:44",{"id":283,"version":284,"summary_zh":285,"released_at":286},100499,"v0.2.5-beta.4","## 🚀 ClawX v0.2.5-beta.4\n\nClawX - Graphical AI Assistant based on OpenClaw\n\n### 📦 Downloads\n\nPlease select the appropriate installer for your operating system and architecture:\n\n#### macOS\n- **Apple Silicon (M1\u002FM2\u002FM3\u002FM4)**: `ClawX-*-mac-arm64.dmg`\n- **Intel (x64)**: `ClawX-*-mac-x64.dmg`\n\n#### Windows\n- **Installer (x64)**: `ClawX-*-win-x64.exe`\n- **Installer (ARM64)**: `ClawX-*-win-arm64.exe`\n\n#### Linux\n- **AppImage (x64)**: `ClawX-*-linux-x86_64.AppImage` (Universal format, recommended)\n- **AppImage (ARM64)**: `ClawX-*-linux-arm64.AppImage`\n- **Debian\u002FUbuntu (x64)**: `ClawX-*-linux-amd64.deb`\n- **Debian\u002FUbuntu (ARM64)**: `ClawX-*-linux-arm64.deb`\n- **RPM (x64)**: `ClawX-*-linux-x86_64.rpm`\n\n### 📝 Release Notes\n\nSee the auto-generated release notes below for detailed changes.\n\n### ⚠️ Installation Notes\n\n- **macOS**: On first launch, you may see \"cannot verify developer\". Go to System Preferences → Security & Privacy to allow the app to run\n- **Windows**: SmartScreen may block the app. Click \"More info\" → \"Run anyway\" to proceed\n- **Linux AppImage**: First run `chmod +x ClawX-*.AppImage` to add execute permission. On Ubuntu 22.04 you may also need `sudo apt install libfuse2`; on Ubuntu 24.04 use `sudo apt install libfuse2t64`\n- **Linux .deb (Ubuntu 24.04)**: If installation fails due to missing dependencies, use `sudo apt install libgtk-3-0t64 libnotify4t64 libxss1t64` before installing\n\n---\n\n💬 Found an issue? Please submit an [Issue](https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fissues)\n\n\n## What's Changed\n* fix token cost by @su8su in https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fpull\u002F579\n\n\n**Full Changelog**: https:\u002F\u002Fgithub.com\u002FValueCell-ai\u002FClawX\u002Fcompare\u002Fv0.2.5-beta.3...v0.2.5-beta.4","2026-03-18T11:14:25"]