No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by daxiondi · Codex Skill · ★ 85
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is openclaw-desktop safe to install? View the security audit →
openclaw-desktop 默认中文文档。English docs: README.en.md 是面向普通用户的 OpenClaw 桌面版，目标是「安装即用、零门槛」。 这个项目的好处 零依赖体感：用户安装一个桌面包即可，不需要先手动装一堆 CLI 和环境。 离线友好：安装包内置 OpenClaw 离线载荷，弱网/无外网场景也能完成初始化。 登录更顺滑：支持 OAuth 登录流程，并可复用本机已有的登录状态（如 Codex）。 国内可用：同时支持 OAuth 路线和 API Key 路线（可接入国内模型/网关）。 官方能力不丢：可直接打开 OpenClaw 官方本地页面使用聊天与配置能力。 跨平台交付：统一产出 macOS / Windows / Linux 安装包。 用户快速开始 打开 Releases 页面下载对应系统安装包。 安装并启动 。 在引导页选择登录方式： OAuth（如 Codex / Claude / Gemini / Qwen Portal） API Key（可对接国内兼容端点） 本地 Ollama 登录完成后即可进入聊天和模型配置。 Windows 离线安装说明 目标：弱网/离线用户只需一次安装，尽量不依赖复杂外网步骤。 当前 Windows 提供两条路径： 自动路径（推荐） 正常安装并启动 。 首次引导时，程序会优先尝试安装包内离线载荷。 如果安装包内离线载荷缺失或不完整，会自动下载并提取 后继续安装。 手动路径（兜底） 在引导页点击 。 选择你已经下载好的 。 程序会自动提取并安装离线载荷，然后继续初始化。 为什么 Windows 还会出现“安装包内离线载荷不可用”： Windows 的打包链路里，资源注入偶发失败/路径不稳定（表现为构建产物里缺少 ）。 为避免用户被阻塞，我们增加了运行时自动下载兜底 + 手动选择 portable 兜底，两条路径都不要求用户手动配环境。 验证 Windows portable 安装（dev/CI） 该脚本用于验证 Windows portable zip 是否能完成离线自举：下载/解压载荷 → 离线安装 → 启动 gateway → 验证本地页面可访问。 沟通交流群 扫码加入 交流群： 在线更新（自动检测 + 一键更新） 应用顶部已内置在线更新入口： 启动后自动静默检测新版本。 检
| Stars | 85 |
| Forks | 22 |
| Language | Rust |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 62.2775840348901/100 |
| Open Issues | 14 |
| Last Updated | 2026-05-06 |
| Created | 2026-02-18 |
| Platforms | cli, rust |
| Est. Tokens | ~839k |
These tools work well together with openclaw-desktop for enhanced workflows:
Looking for a openclaw-desktop alternative? If you're comparing openclaw-desktop with other codex skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Kindly Web Search MCP Server: Web search + robust content retrieval for AI coding tools (Claude Code, Codex, C
🚀 JoySafeter: An enterprise AI Agent Platform—Not just chatting. building、running、testing, and tracing autono
Bringing Human Capabilities to AI Agents
A collection of AI agent skills for Clawdbot, Claude Code, Codex
Curated awesome list for OpenClaw (formerly Moltbot/Clawdbot): skills, plugins, memory systems, MCP tools, dep
Client SDK, Claude plugin and skill framework for the Pinion protocol. x402 micropayments on Base.
Explore other popular codex skill tools:
openclaw-desktop is 零依赖体感：用户安装一个桌面包即可，不需要先手动装一堆 CLI 和环境。 离线友好：安装包内置 OpenClaw 离线载荷，弱网/无外网场景也能完成初始化。 跨平台交付：统一产出 macOS / Windows / Linux 安装包。. It is categorized as a Codex Skill with 85 GitHub stars.
openclaw-desktop is primarily written in Rust. It covers topics such as opeclaw-desktop, openclaw.
You can find installation instructions and usage details in the openclaw-desktop GitHub repository at github.com/daxiondi/openclaw-desktop. The project has 85 stars and 22 forks, indicating an active community.
openclaw-desktop is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to openclaw-desktop on Agent Skills Hub include kindly-web-search-mcp-server, JoySafeter, human-mcp. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: