[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-82784":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":41,"readmeContent":42,"aiSummary":43,"trendingCount":16,"starSnapshotCount":16,"syncStatus":44,"lastSyncTime":45,"discoverSource":46},82784,"wesight","freestylefly\u002Fwesight","freestylefly","Open-source desktop AI agent workspace with one-click Claude Code, Codex, OpenClaw, Hermes Agent setup and custom LLM model routing.","https:\u002F\u002Fwesight.ai",null,"TypeScript",542,142,4,6,0,55,87,372,165,10.47,"MIT License",false,"main",true,[27,28,29,30,31,32,33,34,35,36,37,38,39,40],"agent-workspace","ai-agent","claude-code","codex","desktop-app","electron","hermes-agent","llm","local-first","macos","model-router","openclaw","react","typescript","2026-06-12 02:04:27","# WeSight\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme-banner.svg\" alt=\"WeSight desktop AI agent workspace\" width=\"900\">\n\u003C\u002Fp>\n\n\u003Ch3 align=\"center\">\n  Desktop AI Agent Workspace for Local Coding Agents\n\u003C\u002Fh3>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Fstargazers\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffreestylefly\u002Fwesight?style=flat-square&color=1b79ff\" alt=\"GitHub stars\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Fnetwork\u002Fmembers\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002Ffreestylefly\u002Fwesight?style=flat-square&color=14b8a6\" alt=\"GitHub forks\">\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Ffreestylefly\u002Fwesight?style=flat-square&color=f59e0b\" alt=\"Latest release\">\u003C\u002Fa>\n  \u003Ca href=\"LICENSE\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002Ffreestylefly\u002Fwesight?style=flat-square&color=64748b\" alt=\"MIT License\">\u003C\u002Fa>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fplatform-macOS%20Apple%20Silicon-111827?style=flat-square&logo=apple&logoColor=white\" alt=\"macOS Apple Silicon\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>English\u003C\u002Fstrong> | \u003Ca href=\"README_zh.md\">简体中文\u003C\u002Fa>\n\u003C\u002Fp>\n\nWeSight is an open-source desktop control console for local AI agents. It helps you install or reuse Claude Code, Codex, OpenClaw, Hermes Agent, OpenCode, Qwen Code, DeepSeek-TUI, and the built-in agent runtime, then gives them a visual workspace for chat, tools, files, IM channels, skills, model providers, runtime metrics, and desktop companion workflows.\n\n> Early public releases ship macOS Apple Silicon first. If WeSight helps your agent workflow, a Star makes the project easier for more builders to discover.\n\n## Quick Links\n\n- Website: [wesight.ai](https:\u002F\u002Fwesight.ai\u002F)\n- Latest release: [github.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest](https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest)\n- Screenshots: [Screenshots](#screenshots)\n- Core features: [Core Features](#core-features)\n- Agent engines: [Agent Engines](#agent-engines)\n- Development: [Quick Start](#quick-start)\n\n## Why WeSight\n\nTerminal-native coding agents are powerful, while their setup, model routing, permissions, IM entry points, file changes, and runtime metrics often live in separate places. WeSight turns those moving pieces into one desktop workspace:\n\n- Install, detect, and reuse local agent CLIs from a beginner-friendly UI.\n- Run coding agents through a visual chat with tool panels, slash commands, file diffs, and permission prompts.\n- Connect agent tasks to IM channels such as Feishu, with per-engine configuration.\n- Track every task with engine, model, token usage, TTFT, TPS, tool latency, steps, status, and duration.\n- Extend workflows through SkillHub skills, built-in skills, scheduled tasks, memory, and a desktop pet that follows active work.\n\n## Screenshots\n\n\u003Ctable>\n  \u003Ctr>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Fcowork-chat.png\" alt=\"WeSight Cowork chat\">\n    \u003C\u002Ftd>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Fagent-engines.png\" alt=\"WeSight agent engine settings\">\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd>\u003Cstrong>Cowork Chat\u003C\u002Fstrong>\u003Cbr>Run local coding agents as a desktop chat with engine and model controls.\u003C\u002Ftd>\n    \u003Ctd>\u003Cstrong>Agent Engines\u003C\u002Fstrong>\u003Cbr>Configure Claude Code, Codex, OpenClaw, Hermes Agent, OpenCode, Qwen Code, DeepSeek-TUI, and the built-in runtime.\u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Fruntime-dashboard.png\" alt=\"WeSight runtime dashboard\">\n    \u003C\u002Ftd>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Flive-workspace.png\" alt=\"WeSight live workspace\">\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd>\u003Cstrong>AI Runtime Dashboard\u003C\u002Fstrong>\u003Cbr>Inspect engine, model, tokens, TTFT, output-phase TPS, estimated model TPS, cost, and status.\u003C\u002Ftd>\n    \u003Ctd>\u003Cstrong>Live Workspace\u003C\u002Fstrong>\u003Cbr>Watch file writes, code changes, tool activity, and generated artifacts while the agent works.\u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Fskills-marketplace.png\" alt=\"WeSight skills marketplace\">\n    \u003C\u002Ftd>\n    \u003Ctd width=\"50%\">\n      \u003Cimg src=\"public\u002Freadme\u002Fscreenshots\u002Fstudio-pet.png\" alt=\"WeSight studio and desktop companion\">\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd>\u003Cstrong>Skills Marketplace\u003C\u002Fstrong>\u003Cbr>Browse SkillHub categories, install skills locally, and manage installed skills from WeSight.\u003C\u002Ftd>\n    \u003Ctd>\u003Cstrong>Studio & Pet\u003C\u002Fstrong>\u003Cbr>Use a visual office-style workspace and desktop companion to follow active agent tasks.\u003C\u002Ftd>\n  \u003C\u002Ftr>\n\u003C\u002Ftable>\n\n## Core Features\n\n- **Agent Engines** - Run Claude Code, Codex, OpenClaw, Hermes Agent, OpenCode, Qwen Code, DeepSeek-TUI, or the built-in runtime from the same workspace.\n- **One-click setup** - On macOS, WeSight can install supported local CLIs or detect the ones already present on the machine.\n- **Unified model providers** - Configure official OpenAI, Anthropic Claude, Google Gemini, DeepSeek, Qwen, Moonshot, Ollama, OpenRouter, GitHub Copilot, and custom OpenAI-compatible endpoints.\n- **Local CLI configuration** - Use existing Claude Code, Codex, OpenClaw, Hermes Agent, OpenCode, Qwen Code, or DeepSeek-TUI accounts and config when you already have a working terminal setup.\n- **Graphical tool execution** - View commands, files, permissions, slash commands, outputs, generated images, and tool results inside the chat flow.\n- **IM Agent Hub** - Route Feishu messages into OpenClaw, Hermes Agent, Claude Code, or Codex, with per-engine bot profiles.\n- **AI Runtime Dashboard** - Measure calls by engine, model, source, status, tokens, completion time, TTFT, output-phase TPS, estimated model TPS, tool latency, and agent steps.\n- **Live Workspace** - Open a right-side workspace for live code writing, static diffs, runtime monitoring, task todos, skills, and artifacts.\n- **SkillHub Marketplace** - Discover, categorize, install, enable, disable, update, and remove local WeSight skills.\n- **Scheduled Tasks** - Create recurring agent jobs for research, reports, monitoring, inbox cleanup, and reminders.\n- **Memory and personalization** - Extract useful preferences from conversations and reuse them across future sessions.\n- **Desktop pet and studio** - Keep a lightweight desktop companion and a pixel-style studio view for active tasks.\n\n## ❤️ Sponsors\n\n> [Want to appear here?](public\u002Freadme\u002Fcommunity\u002Fwechat-personal.jpg) Add me on WeChat and include your product name plus a short project sponsorship note in the friend request.\n\n| Sponsor | Description |\n| ------- | ----------- |\n| \u003Ca href=\"https:\u002F\u002Fpptoken.cc\u002F\">\u003Cimg src=\"public\u002Freadme\u002Fsponsors\u002Fpptoken.png\" alt=\"PPToken\" width=\"240\">\u003C\u002Fa> | Project sponsor. PPToken provides API relay and key distribution for ChatGPT, Claude, Gemini and other mainstream AI models, with low latency, high availability, pay-as-you-go billing, and flexible subscription plans. |\n| \u003Ca href=\"https:\u002F\u002Fciyuan.today\u002F\">\u003Cimg src=\"public\u002Freadme\u002Fsponsors\u002Fciyuan-api.jpg\" alt=\"Ciyuan API\" width=\"240\">\u003C\u002Fa> | Project sponsor. Ciyuan API aims to become a one-stop AI interface platform for developers, providing stable, low-latency, and highly available large model API services to make AI application development simpler. |\n\n## Agent Engines\n\n| Engine           | Best For                                                    | Setup Path                                      |\n| ---------------- | ----------------------------------------------------------- | ----------------------------------------------- |\n| Built-in runtime | General desktop cowork sessions and skills                  | Included in WeSight                             |\n| Claude Code      | Claude Code workflows with a graphical chat surface         | One-click install or existing local CLI config  |\n| Codex            | Codex CLI workflows, local task execution, and IM control   | One-click install or existing local CLI config  |\n| OpenClaw         | Runtime gateway, IM channels, sandbox-style agent work      | Local runtime\u002FCLI reuse or WeSight setup flow   |\n| Hermes Agent     | Local Hermes Agent gateway and IM-style runtime experiments | Official installer or existing local CLI config |\n| OpenCode         | OpenCode terminal agent workflows                           | One-click install or existing local CLI config  |\n| Qwen Code        | Qwen-friendly coding workflows and DashScope setups         | One-click install or existing local CLI config  |\n| DeepSeek-TUI     | DeepSeek-TUI HTTP\u002FSSE runtime and tool streaming            | One-click install or existing local CLI config  |\n\n## Model Providers\n\nWeSight keeps model setup in one place, then maps it into the selected engine when that engine follows WeSight settings.\n\n- Add multiple providers and models.\n- Use official OpenAI, Anthropic Claude, and Google Gemini providers.\n- Add OpenAI-compatible providers for DeepSeek, Qwen, Moonshot, Ollama, OpenRouter, GitHub Copilot, local gateways, or private endpoints.\n- Switch between WeSight-managed model settings and existing local CLI configuration.\n- Import or sync local engine configuration when you want WeSight to manage it.\n\n## Download\n\nPublic desktop builds are published through GitHub Releases:\n\n- Website: [wesight.ai](https:\u002F\u002Fwesight.ai\u002F)\n- Latest release: [github.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest](https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest)\n\nEarly public releases currently ship macOS Apple Silicon builds first. Release assets are intended for end users. CI artifacts are short-lived build outputs for maintainers to test before a release is published.\n\n## Download And Install\n\n### 1. Download the DMG\n\nDownload `WeSight-*-arm64.dmg` from the [latest release](https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight\u002Freleases\u002Flatest), open it, and drag `WeSight.app` into the `Applications` folder.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme\u002Ftutorial\u002Finstall-dmg.svg\" alt=\"WeSight DMG install guide\" width=\"760\">\n\u003C\u002Fp>\n\n### 2. If macOS says the app is damaged\n\nPreview builds are not signed and notarized yet. macOS may show a message like:\n\n> “WeSight.app” is damaged and cannot be opened. You should move it to the Trash.\n\nThis is usually a Gatekeeper quarantine warning for an unsigned app. It does not mean the downloaded package is corrupted. Click Cancel first.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme\u002Ftutorial\u002Fmacos-damaged-warning.svg\" alt=\"macOS unsigned app warning\" width=\"620\">\n\u003C\u002Fp>\n\n### 3. Remove the quarantine attribute\n\nOpen the built-in macOS Terminal app and run:\n\n```bash\nxattr -cr \u002FApplications\u002FWeSight.app\n```\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme\u002Ftutorial\u002Fxattr-terminal.svg\" alt=\"Run xattr in Terminal\" width=\"760\">\n\u003C\u002Fp>\n\nAfter the command finishes, open WeSight again. If you installed WeSight somewhere else, replace `\u002FApplications\u002FWeSight.app` with the actual path to your `WeSight.app`.\n\n## Quick Start\n\n### Requirements\n\n- Node.js `>=24 \u003C25`\n- npm\n\n### Development\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Ffreestylefly\u002Fwesight.git\ncd wesight\nnpm install\nnpm run electron:dev\n```\n\nThe Vite dev server runs at `http:\u002F\u002Flocalhost:5175`.\n\n### Development With Agent Runtimes\n\n```bash\n# Start WeSight and detect supported local agent CLIs from Settings\nnpm run electron:dev\n\n# Convenience aliases currently point to the same development entry\nnpm run electron:dev:openclaw\nnpm run electron:dev:hermes\n```\n\nUseful OpenClaw development variables:\n\n```bash\n# Override OpenClaw source location\nOPENCLAW_SRC=\u002Fpath\u002Fto\u002Fopenclaw npm run electron:dev:openclaw\n\n# Force OpenClaw runtime rebuild\nOPENCLAW_FORCE_BUILD=1 npm run electron:dev:openclaw\n\n# Skip OpenClaw version checkout for local OpenClaw development\nOPENCLAW_SKIP_ENSURE=1 npm run electron:dev:openclaw\n```\n\n## Build\n\n```bash\n# TypeScript + Vite\nnpm run build\n\n# Electron main process\nnpm run compile:electron\n\n# ESLint\nnpm run lint\n```\n\n## Packaging\n\n```bash\n# macOS\nnpm run dist:mac\nnpm run dist:mac:x64\nnpm run dist:mac:arm64\nnpm run dist:mac:universal\n\n# Windows\nnpm run dist:win\n\n# Linux\nnpm run dist:linux\n```\n\nManaged runtime metadata is declared in `package.json`. Generated runtime folders, build artifacts, local secrets, and packaged release output are ignored by Git.\n\n## Architecture\n\nWeSight uses Electron process isolation. The renderer never directly accesses Node.js APIs; privileged operations go through a typed preload bridge and IPC handlers in the main process.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme-architecture.svg\" alt=\"WeSight architecture principle diagram\" width=\"960\">\n\u003C\u002Fp>\n\n### Main Process\n\n- Window lifecycle, tray behavior, desktop pet windows, and deep links\n- SQLite persistence for settings, sessions, messages, runtime calls, skills, and auth tokens\n- Agent engine routing and external CLI adapters\n- OpenClaw and Hermes gateway lifecycle helpers\n- IM gateway integrations and native Feishu routing\n- Skill management, scheduled tasks, file activity tracking, and runtime telemetry\n\n### Renderer\n\n- React + Redux Toolkit + Tailwind CSS\n- Cowork chat UI, studio view, live workspace, runtime dashboard, and artifacts\n- Engine selector, model selector, settings, skills, MCP, agents, IM, memory, and appearance UI\n- Stream rendering for messages, tool calls, command output, slash command panels, files, images, and permission prompts\n\n### Key Directories\n\n```text\nsrc\u002Fmain\u002F\n  main.ts                         Electron entry and IPC handlers\n  preload.ts                      Safe renderer bridge\n  sqliteStore.ts                  Local persistence\n  coworkStore.ts                  Session and message storage\n  libs\u002FagentEngine\u002F               Engine adapters and router\n  libs\u002FexternalAgent*.ts          External CLI setup and config helpers\n  im\u002F                             IM gateway integrations\n\nsrc\u002Frenderer\u002F\n  App.tsx                         App shell\n  components\u002Fcowork\u002F              Chat, studio, activity workspace, engine UI\n  components\u002FSettings.tsx         Model, engine, IM, skills, memory, and app settings\n  components\u002Fpet\u002F                 Desktop companion UI\n  services\u002F                       IPC wrappers and app services\n  store\u002Fslices\u002F                   Redux state\n\nSKILLs\u002F                           Built-in skills\nscripts\u002F                          Runtime, packaging, and setup scripts\nsrc\u002Fshared\u002F                       Shared constants and types\n```\n\n## Built-in Skills\n\nWeSight includes a broad skills library for day-to-day agent work and connects to SkillHub for marketplace installation.\n\n| Area           | Examples                                                                  |\n| -------------- | ------------------------------------------------------------------------- |\n| Research       | web search, tech news, stock research, film\u002Fmusic search                  |\n| Documents      | DOCX, XLSX, PPTX, PDF processing                                          |\n| Automation     | Playwright, local tools, scheduled tasks                                  |\n| Creative       | Remotion video, frontend design, canvas design, image and video workflows |\n| Communication  | IMAP\u002FSMTP email and IM channels                                           |\n| Agent building | skill creator, skill vetting, custom planning                             |\n\nSkills can be installed, enabled, disabled, deleted, and routed from the desktop UI.\n\n## Security Model\n\n- Context isolation is enabled.\n- Node integration is disabled in the renderer.\n- Sensitive operations run through main-process IPC.\n- Tool execution can surface permission requests before running.\n- Local data is stored in SQLite under the app data directory.\n- Runtime folders, build artifacts, generated assets, and local secrets are ignored by Git.\n\n## Roadmap Ideas\n\n- More engine adapters and runtime profiles\n- Better local configuration import and provider sync flows\n- Richer IM agent profiles and message formats\n- Shareable task templates\n- More visual inspection tools for long-running agent tasks\n- Skill marketplace updates, reviews, and version management\n\n## Community WeChat Group\n\nScan the QR code below to join the WeSight WeChat group and talk with other builders. The QR code is valid until June 8, 2026; if it expires, follow the official account below to get the latest invite.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Freadme\u002Fcommunity\u002Fwechat-group.jpg\" alt=\"WeSight WeChat Group\" width=\"320\">\n\u003C\u002Fp>\n\n## Star History\n\n[![Star History Chart](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=freestylefly\u002Fwesight&type=Date)](https:\u002F\u002Fstar-history.com\u002F#freestylefly\u002Fwesight&Date)\n\n## WeChat Official Account\n\nSearch **苍何** on WeChat or scan the QR code below to follow Canghe's original WeChat official account. Reply with **AI** to get more AI prompt and agent workflow resources.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"public\u002Fwechat-official-account.png\" alt=\"Canghe WeChat Official Account\" width=\"280\">\n\u003C\u002Fp>\n\n## Acknowledgements\n\nWeSight is shaped by many excellent open-source projects and agent community practices. Special thanks to:\n\n- [LobsterAI](https:\u002F\u002Fgithub.com\u002Fnetease-youdao\u002FLobsterAI) for inspiration around desktop personal assistant agents, README structure, and open-source launch practice.\n- [OpenClaw](https:\u002F\u002Fgithub.com\u002Fopenclaw\u002Fopenclaw) for exploring local agent runtimes, gateways, and IM agent capabilities.\n- [Hermes Agent](https:\u002F\u002Fgithub.com\u002FNousResearch\u002Fhermes-agent) for local agent runtime, gateway, and model configuration ideas.\n- [Star-Office-UI](https:\u002F\u002Fgithub.com\u002Fringhyacinth\u002FStar-Office-UI) for the pixel-style AI studio inspiration.\n- [SkillHub](https:\u002F\u002Fskillhub.lol\u002Fskills) for ideas around skill discovery, installation, and marketplace flows.\n- The terminal-agent ecosystem around Claude Code, Codex, OpenCode, Qwen Code, DeepSeek-TUI, and the builders pushing local AI agent workflows forward.\n\n## License\n\nMIT. See [LICENSE](LICENSE).\n","WeSight 是一个开源的桌面AI代理工作区，支持一键安装和配置Claude Code、Codex、OpenClaw、Hermes Agent等本地编码代理，并提供自定义大语言模型路由功能。该项目使用TypeScript编写，基于Electron框架构建，具备直观的用户界面，能够帮助用户轻松管理多个AI代理及其相关任务。它允许通过图形化聊天界面与工具面板进行交互，支持文件差异查看、权限请求等功能；同时还可以将代理任务连接到飞书等即时通讯渠道，并跟踪每个任务的详细运行指标。适用于需要在本地环境中高效管理和利用各种AI编码助手的开发者或团队。",2,"2026-06-11 04:09:13","CREATED_QUERY"]