Maintainer of this skill?
by DLYZZT
A local-first, cross-platform Electron desktop workspace for Pi Coding Agent, with sessions, project files, browser tools, skills, plugins, and messaging integrations.
# Add to your Claude Code skills
git clone https://github.com/DLYZZT/pi-desktopGuides for using ai agents skills like pi-desktop.
Last scanned: 8/7/2026
{
"issues": [
{
"type": "npm-audit",
"message": "@earendil-works/pi-coding-agent: Vulnerability found",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "@electron/node-gyp: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "@electron/rebuild: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "app-builder-lib: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "brace-expansion: brace-expansion: DoS via exponential-time expansion of consecutive non-expanding {} groups",
"severity": "high"
},
{
"type": "npm-audit",
"message": "builder-util: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "builder-util-runtime: electron-updater: Cross-origin redirect leaks `PRIVATE-TOKEN` and mixed-case `Authorization` credentials in `builder-util-runtime`",
"severity": "high"
},
{
"type": "npm-audit",
"message": "cacache: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "dmg-builder: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "dompurify: DOMPurify: `CUSTOM_ELEMENT_HANDLING` bypasses `afterSanitizeElements` for allowed custom elements.",
"severity": "low"
},
{
"type": "npm-audit",
"message": "electron-builder: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "electron-builder-squirrel-windows: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "electron-publish: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "esbuild: esbuild allows arbitrary file read when running the development server on Windows",
"severity": "low"
},
{
"type": "npm-audit",
"message": "ip-address: ip-address: Address4 decodes leading-zero octets as decimal while resolvers decode them as octal, allowing SSRF and trust-boundary bypass",
"severity": "high"
},
{
"type": "npm-audit",
"message": "js-yaml: JS-YAML: Quadratic CPU consumption in !!omap resolution (3.x and 4.x) — CVE-2026-59870 fix not backported",
"severity": "high"
},
{
"type": "npm-audit",
"message": "make-fetch-happen: Vulnerability found",
"severity": "high"
},
{
"type": "npm-audit",
"message": "mermaid: Mermaid configuration APIs allow prototype pollution",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "postcss: PostCSS: Path Traversal in Previous Source Map Auto-Loading (sourceMappingURL) leads to Arbitrary .map File Disclosure",
"severity": "high"
},
{
"type": "npm-audit",
"message": "protobufjs: protobufjs: Denial of Service via infinite loop in .proto option parsing",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "tar: node-tar Vulnerable to Arbitrary File Creation/Overwrite via Hardlink Path Traversal",
"severity": "critical"
},
{
"type": "npm-audit",
"message": "undici: undici vulnerable to downstream response desynchronization via retry interceptor",
"severity": "high"
}
],
"status": "FAILED",
"scannedAt": "2026-08-07T05:40:30.231Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how pi-desktop compares with popular alternatives.
pi-desktop is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by DLYZZT. A local-first, cross-platform Electron desktop workspace for Pi Coding Agent, with sessions, project files, browser tools, skills, plugins, and messaging integrations. It has 321 GitHub stars.
pi-desktop failed SkillsLLM's automated security scan, which flagged one or more high-severity issues. Review the Security Report section carefully before using it.
Clone the repository with "git clone https://github.com/DLYZZT/pi-desktop" and add it to your Claude Code skills directory (see the Installation section above).
pi-desktop is primarily written in TypeScript. It is open-source under DLYZZT 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 pi-desktop against similar tools.
No comments yet. Be the first to share your thoughts!
Requires a passing catalog security scan.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
把 Pi Coding Agent 变成真正的桌面工作台。
本地优先 · 零内部服务器 · 跨平台应用
English · 简体中文
下载 v0.4.3 · 截图 · 功能 · 快速开始 · 架构 · 变更记录 · 路线图
主工作区与主要设置界面，点击图片可查看原图。
@ 文件引用tool_history_get 查询原文WebContentsView 承载真实 Chromium 页面，支持多 Tab、临时/持久 Profile、登录态、下载、上传和代理process_* tools 持续运行 Vite、React、Three.js、Storybook、Flask、Spring Boot、mock API 和 watch build；普通短命令继续走 Bash&、nohup 或外部终端受管后台进程支持 macOS、Linux 和 Windows 11 x64；Windows ARM64、Windows Server 和 32 位 Windows 暂不支持。
agent.stop，因此 Agent close 会明确关闭其所在 pane，不会静默伪造停止语义.docx）文档max 推理级别开发分支的模型配置可先运行 npm run check:models-config 只读预检；无效字段会使整份自定义配置暂不生效，设置页保留原内容以便修正。OpenRouter 的个性化目录与 Pi CLI 共享缓存，不按 key 或 endpoint 分区；不同凭据交替使用、特别是离线启动时，列表可能来自另一个运行环境，不能将缓存目录视为当前 key 已获授权。
在“设置 → 模型 → 自动路由”启用 Auto，选择快模型、强模型及各自的思考级别，保存后在聊天模型菜单中选择 Auto。
high 及以上使用强模型，其余使用快模型；也可配置决策模型，根据每次新提问的复杂度选择模型路由配置在新建、重新加载会话或重新选择 Auto 后生效。
通过会话工具菜单启用 Codemode（“完全访问”预设已包含），让 Agent 按需生成图片，或调用决策模型进行分类、是非判断和评分。支持图片输入的分类器也可判断截图和图片。模型按提供商的现有凭据发现，分类器作为辅助能力使用，不进入聊天主模型选择器。
每次模型调用都遵循 Codemode 权限，支持停止，并保存原始输入、结果和执行状态；用量计入会话统计，可在已有工具记录中查看。OpenAI 决策接口需要 API key，ChatGPT 订阅登录不提供该能力。TypeSafe Jev 可在“设置 → 模型 → 添加服务商”中搜索 TypeSafe 或 Jev，填写 API key 后使用；详情页会列出 jev-latest。可以让 Agent“使用可用的决策模型给这些条目分类”。
生成的图片可在最终回复及重开的会话历史中查看。服务商详情页列出可用的图片与决策模型，并提供测试请求；测试使用所配置的服务商凭据，可能产生用量费用。
/help、/status、/new、/compact 和 /reloadsandbox: true、严格 CSP 与类型化 IPC 契约最新稳定版为 v0.4.3，提供 macOS Apple Silicon / Intel、Windows x64 和 Linux x64 安装包。
v0.4.3 内置 Pi Coding Agent 1.4.0，改进上下文压缩、工具调用与多会话代码编排，补充模型配置兼容提示，并精简普通服务商的 API Key 配置入口。也可按下方步骤从源码运行。
普通用户使用 Agent 本身无需单独安装 Pi CLI、Pi Coding Agent、Node.js 或 npm；安装桌面应用并配置模型提供商后即可使用。Skills、Plugins 或 Agent 脚本需要额外开发工具时，应用会优先复用健康的系统安装，也可以在用户确认后安装应用私有运行时。
应用会读取 ~/.pi/agent/ 中的会话与配置。如果你已经使用 Pi CLI，可以直接复用现有数据，无需迁移；此前没有使用过 Pi CLI 也不影响使用。
Pi Desktop 会先发现并验证用户已经安装的 Node.js/npm、Python、Git、Bash、uv、jq 和 Bun；内置的 rg/fd 保证离线搜索可用。
>=22.19.0 <23）git clone https://github.com/DLYZZT/pi-desktop.git
cd pi-desktop
npm ci
npm run dev
Pi Agent Desktop 使用 Electron 三进程模型，将高权限桌面能力、Agent 运行时和 UI 隔离开来。
flowchart LR
Main["Electron Main<br/>窗口 · 托盘 · 协议 · Host 监督"]
Host["Agent Host / utilityProcess<br/>Pi Agent · 会话 · 文件 · 配置"]
UI["Renderer<br/>React 19 · Vite"]
Browser["Main-owned WebContentsView<br/>远程网页 · Profile · 网络策略"]
Processes["受管项目进程<br/>dev server · watcher · mock API"]
Data["~/.pi/agent/<br/>会话 · 模型 · 配置"]
Main --> Host
Main --> UI
Main --> Browser
Host -->|"revisioned Browser RPC"| Main
Host -->|"POSIX worker / Windows Job helper