[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-93888":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":40,"readmeContent":41,"aiSummary":42,"trendingCount":16,"starSnapshotCount":16,"syncStatus":43,"lastSyncTime":44,"discoverSource":45},93888,"cindy","makecindy\u002Fcindy","makecindy","Consider it done. The open-source AI agent that works out of the box · 想到，就能做到。开源、开箱即用的 AI Agent。","https:\u002F\u002Fcindy.app",null,"TypeScript",8078,943,29,1759,0,704,4597,4824,4815,114.92,"Apache License 2.0",false,"main",true,[27,28,29,30,31,32,33,34,35,36,37,38,39],"agent","ai-agent","ai-assistant","android","claude-code","codex","electron","ios","llm","macos","react-native","typescript","windows","2026-10-29 08:00:02","\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fcindy.app\">\u003Cimg src=\".github\u002Fassets\u002Fhero-en.webp\" width=\"100%\" alt=\"CINDY — Consider it done. The open-source AI agent that works out of the box, doing real work on your own computer.\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>English\u003C\u002Fstrong> · \u003Ca href=\"README.zh-CN.md\">简体中文\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"LICENSE\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-Apache--2.0-blue.svg\" alt=\"License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmakecindy\u002Fcindy\u002Factions\u002Fworkflows\u002Fci.yml\">\u003Cimg src=\"https:\u002F\u002Fgithub.com\u002Fmakecindy\u002Fcindy\u002Factions\u002Fworkflows\u002Fci.yml\u002Fbadge.svg\" alt=\"CI\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fnodejs.org\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fnode-22.x-brightgreen.svg\" alt=\"Node.js 22.x\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fpnpm.io\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpnpm-10-orange.svg\" alt=\"pnpm\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  🌐 \u003Ca href=\"https:\u002F\u002Fcindy.app\">cindy.app\u003C\u002Fa>  ·  ⬇️ \u003Ca href=\"https:\u002F\u002Fcindy.app\u002Fdownload\u002F\">Download\u003C\u002Fa>\n\u003C\u002Fp>\n\n\nCindy is an open-source AI agent that works out of the box. She brings multiple\nharnesses, models and tools into one agent that finishes real work in your\nprojects and apps. Ready from day one, yours to shape over time.\n\nCindy runs locally on your own machine, using your real files and logged-in\napps. The first supported harnesses are **Claude Code** and **Codex** — more are\nbeing added, and a native harness is in the works. Models and harnesses mix\nfreely and can switch mid-task while your workspace, memory, skills and tools\nstay continuous; one task can even be planned, executed in parallel, and\nreviewed by agents on different harness × model combos. She can drive your\nbrowser, computer and phone, and take work from IM and schedules.\n\nThis repository is the open-source **client** for Cindy — the desktop and mobile\napps plus their shared packages, organized as a pnpm monorepo.\n\nThe client is free to use, and its source code is open under Apache-2.0. Bring\nmodels your way: sign in to the official Cindy service (usage deducted\ntransparently), authorize the **Claude Code \u002F Codex Coding Plan** you already\npay for and keep using it inside Cindy — no duplicate bill — connect your own\nAPI keys, or use local models.\nSee [cindy.app](https:\u002F\u002Fcindy.app) for service details,\n[pricing](https:\u002F\u002Fcindy.app\u002F#pricing), and [downloads](https:\u002F\u002Fcindy.app\u002Fdownload\u002F).\n\n## Yours to shape\n\nOpen source means more than visible — it means changeable:\n\n- **Memory** — correct her once and she does it right from then on, shared across harnesses.\n- **Skills** — teach a way of working once and reuse it everywhere; handing them to your team is in the making.\n- **Automation** — recurring work schedules itself, runs itself, reports back.\n- **MCP** — wire your internal tools and business systems into her reach.\n- **Plugins** — reshape features, UI and interactions, shared through an open marketplace *(in the making)*.\n- **Source** — audit, fork, extend, and contribute improvements back under Apache-2.0.\n\nReady out of the box, never boxed in — start with\n[`CONTRIBUTING.en.md`](CONTRIBUTING.en.md) and build Cindy with us.\n\n## What's in this repo\n\n| Path | Description |\n| --- | --- |\n| `apps\u002Fdesktop` | Electron desktop client |\n| `apps\u002Fmobile` | Expo \u002F React Native mobile client |\n| `packages\u002F*` | Shared client capabilities (auth, device-link, agent orchestration, model providers, …) |\n| `apps\u002F*-bin` | Tool binaries shipped with the desktop app; none are committed — claude-code, codex, and ripgrep are downloaded per platform by `pnpm install`, and the Android platform-tools binaries are fetched (pinned version, sha256-verified) before Windows packaging |\n| `cindy-protocol\u002F` | Wire protocol shared with the server (git submodule) |\n\n**Not in this repo:** the backend service lives in a separate\nrepository and is not part of this monorepo.\n\n| Mode | Account requirement | Availability |\n| --- | --- | --- |\n| Hosted service | Cindy cloud account | Use Cindy's full hosted service. See [pricing](https:\u002F\u002Fcindy.app\u002F#pricing). |\n| Local mode | No Cindy sign-in required | Choose “Local mode” on the login screen to use local agents. Server-backed capabilities are unavailable in this mode. |\n\n## Prerequisites\n\n- **Node.js** 22.x\n- **pnpm** 10.x (v11 is not yet supported)\n- **Git LFS**\n\n## Getting started\n\nContributor setup, public submodule initialization, Git LFS, dependency updates,\nand access requirements are maintained in\n[`CONTRIBUTING.en.md`](CONTRIBUTING.en.md).\nThe public checkout only needs the protocol submodule; plugins are installed through\nSkillHub or manually.\n\nMinimal entry point:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fmakecindy\u002Fcindy.git\ncd cindy\ngit submodule update --init --recursive cindy-protocol\ngit lfs pull\npnpm install\n```\n\n## Development entry points\n\n```bash\n# Mainland China Cindy account\npnpm restart:desktop:remote --region=cn\n\n# Global Cindy account\npnpm restart:desktop:remote --region=global\n```\n\nRemote development uses your own Cindy cloud account and existing login state, so\nyou can continue existing sessions and work while developing the client. Use `cn`\nfor Mainland China accounts and `global` for everyone else; do not rely\non the internal default. Full desktop, mobile, data-isolation, and validation\nworkflows are in [`CONTRIBUTING.en.md`](CONTRIBUTING.en.md).\n\n“Local mode” on the login screen is an unauthenticated local-agent mode, not a\nconnection to a local server. Server-backed capabilities are unavailable in this\nmode.\n\n**About the default servers:** the client connects to Cindy's official cloud\nservices by default (endpoint manifests in\n[`config\u002Fendpoint.json`](config\u002Fendpoint.json) and\n[`config\u002Fendpoint.global.json`](config\u002Fendpoint.global.json); desktop\nauto-updates also come from the official CDN). This is intentional — external\ndevelopers don't need to self-host a server: sign in with your own Cindy\naccount in a dev build and develop \u002F test directly against the official\nservers.\n\n## Architecture\n\n- [`DESIGN.md`](DESIGN.md) — visual design system, color tokens, and UI conventions\n- [`docs\u002FREADME.md`](docs\u002FREADME.md) — complete documentation and rules index\n- [`CONTRIBUTING.en.md`](CONTRIBUTING.en.md) — contributor setup, validation, and submission workflow\n- [`AGENTS.md`](AGENTS.md) — engineering rules, launch\u002Fruntime contracts, and module boundaries\n- [`docs\u002Fdev-rules\u002F`](docs\u002Fdev-rules\u002F) — deep-dive architecture docs (e.g. Orca multi-agent orchestration)\n\n## Contributing\n\nContributions go through pull requests into `main`. Read\n[`CONTRIBUTING.en.md`](CONTRIBUTING.en.md) first, then use\n[`.github\u002FPULL_REQUEST_TEMPLATE.md`](.github\u002FPULL_REQUEST_TEMPLATE.md).\nEvery commit needs a [Developer Certificate of Origin](DCO) sign-off\n(`git commit -s`); a DCO check on each pull request enforces it, and no CLA is\nrequired.\nPlease also follow [`CODE_OF_CONDUCT.en.md`](CODE_OF_CONDUCT.en.md). For ordinary\nusage questions, see [`SUPPORT.en.md`](SUPPORT.en.md); report security issues\nprivately through [`SECURITY.en.md`](SECURITY.en.md).\n\n## Security\n\nNever commit credentials or authorization files to the working tree. If you\ndiscover a security issue, follow [`SECURITY.en.md`](SECURITY.en.md) to report it\nprivately rather than opening a public issue.\n\n## Privacy & telemetry\n\n**Official distribution builds** include [TapDB](https:\u002F\u002Fwww.taptap.io\u002Ftapdb)\nusage analytics for product-level aggregate statistics (device \u002F OS \u002F app-version\nmetadata; associated with your account ID after sign-in). It does **not** collect\nchat content, file content, or working-directory data. In addition, while signed\nin to a cloud account the client sends an online heartbeat to Cindy services\n(account ID, platform, and version only). Crash dumps stay on the local machine\nand are never uploaded automatically.\n\n**Building from source?** You are not required to keep analytics:\n\n- Mobile is off by default — without TapDB credentials (`clientId` \u002F\n  `clientToken`) injected at build time, `apps\u002Fmobile\u002Fsrc\u002Fanalytics\u002FmobileTapdb.ts`\n  is a no-op;\n- Desktop can be fully stripped by removing the `initTapdb()` call in\n  `apps\u002Fdesktop\u002Fsrc\u002Frenderer\u002Findex.tsx` (implementation lives in\n  `apps\u002Fdesktop\u002Fsrc\u002Frenderer\u002Fanalytics\u002F`).\n\n## License \u002F 许可证\n\nExcept as otherwise noted, the source code in this repository is licensed under\nthe [Apache License, Version 2.0](LICENSE). Individual source files do not\ncarry per-file license headers; the repository-root `LICENSE` governs.\n\nModel weights, datasets, prompts, trademarks, and other separately identified\nmaterials may be subject to their own license terms and are not automatically\ncovered by the repository-level Apache-2.0 grant. Third-party open-source\ncomponents retain their own copyright and license. Their attribution notices and\nSPDX SBOMs are managed under [`docs\u002Flegal\u002F`](docs\u002Flegal\u002F), with artifact-specific\noutputs in [`docs\u002Flegal\u002Fnotices\u002F`](docs\u002Flegal\u002Fnotices\u002F). See [`NOTICE`](NOTICE)\nfor this project's copyright and attribution information.\n","Cindy 是一个开源、开箱即用的本地化 AI Agent 客户端，支持跨平台（Windows\u002FmacOS\u002FiOS\u002FAndroid）运行，可直接操作用户本地文件、已登录应用及设备。其核心特点包括多模型与多执行器（Harness）动态协同（如 Claude Code、Codex）、任务级上下文连续性、浏览器\u002F电脑\u002F手机自动化控制、IM 与日程集成，以及可扩展的记忆、技能、MCP 接口和插件机制。适用于个人知识管理、日常办公自动化、开发者辅助编码及企业内部工具链集成等需强隐私保障与本地执行能力的场景。",2,"2026-10-29 03:57:26","CREATED_QUERY"]