by shiwenwen
🦭 会记忆、能持续推进目标、会动态编排多 Agent 的跨端桌面 AI 助手，也可服务化常驻 NAS / 云端 | A cross-device desktop AI agent with memory, autonomous goals, dynamic workflows, and headless deployment
# Add to your Claude Code skills
git clone https://github.com/shiwenwen/hope-agentLast scanned: 5/14/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-14T06:48:06.290Z",
"semgrepRan": false,
"npmAuditRan": true,
"pipAuditRan": true
}hope-agent is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by shiwenwen. 🦭 会记忆、能持续推进目标、会动态编排多 Agent 的跨端桌面 AI 助手，也可服务化常驻 NAS / 云端 | A cross-device desktop AI agent with memory, autonomous goals, dynamic workflows, and headless deployment. It has 1,174 GitHub stars.
Yes. hope-agent passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/shiwenwen/hope-agent" and add it to your Claude Code skills directory (see the Installation section above).
hope-agent is primarily written in Rust. It is open-source under shiwenwen on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh hope-agent against similar tools.
No comments yet. Be the first to share your thoughts!
Hope Agent 想让 AI 助手真正打开就能用，并且长期稳定地陪你工作：一个原生安装包，内置主流模型 Provider 配置模板；填完 API Key 就能开聊，不必先装运行时、学命令行，也不用维护一堆复杂配置。桌面端是最顺手的入口，我们把 GUI、性能、交互细节和系统深度融合放在最前面；在你授权后，它还能观察并控制本机电脑（暂时仅 macOS）。对于复杂或长期任务，你可以只给出最终目标，让 Agent 持续推进；它也能自主决定何时动态编排 Workflow、并行调度多个子 Agent，并通过 Loop 在合适的时间或事件到来时继续工作。整个过程可观察、可暂停、可恢复，审批与完成证据不会因为自动化而被绕过。同时，Hope Agent 也可以作为 HTTP/WS 服务常驻 NAS、自家服务器或云主机，接入 IM 渠道、对接 IDE（ACP），让同一份会话在桌面、浏览器和聊天之间随手交接。随着日常使用，它会持续积累跨会话记忆，在空闲时整理重要内容，把做过的事沉淀成可复用技能。目标很简单：降低使用和维护成本，让日常场景顺手，让长期运行可靠。
从输入框直接选择「目标」「工作流」或「持续推进」，也可以使用 /goal、/workflow、/loop。这些能力面向通用任务：编码、调研、内容创作、资料整理、运营和连接器任务都可以使用；代码 Worktree、Review、LSP 和验证只在编码场景按需叠加。
心智模型：Goal 定义最终要达成什么，Workflow 负责一次具体执行，Loop 决定何时再次推进，Task 呈现当前进度，Mode 控制自主执行强度。它们可以组合使用，也可以各自独立。实现细节见 Goal、Workflow 和 Loop 架构文档。