隐私与广告选择
Git-Stars 会使用必要存储来保障网站运行。可选分析和广告测量脚本默认不加载，只有在你同意后，Google 等合作伙伴才可能按要求使用 Cookie 或类似标识符。 隐私政策
Give your agent a computer 👾
这个仓库详情页已经具备足够的原创分析、来源证据和同类对比，可以作为 Git-Stars 的可索引评测页。
100
评测分
解决的问题
It solves the problem of giving AI agents a persistent, stateful filesystem and execution environment that can be accessed across different runtimes. Unlike traditional approaches that require separate storage and compute, it unifies state and execution in a single Durable Object, enabling seamless synchronization and multiple execution backends without additional infrastructure.
部署现实
当前可用的安装信号是：npm install @cloudflare/computer。这只能作为起点，正式使用前建议让 Codex 或 Claude 检查 README、环境变量、运行时版本和部署目标。
商业可用性
当前记录的许可证是 MIT。这是重要的商业使用信号，但仍应核验许可证全文、依赖许可证、模型/API 条款和商标限制。
能力上限
Developers can build AI agents that perform file operations, run shell commands, execute JavaScript modules, and generate artifacts like PDFs or images. Real-world use cases include automated document generation, code project scaffolding, and interactive chat agents with a working directory. The ceiling is high: it supports full Linux userland in containers, isolated shell environments, and JavaScript execution, all backed by a durable filesystem, enabling complex agent workflows that require persistent state and diverse execution capabilities.
Git-Stars 是独立产品，不隶属于 GitHub 或该项目。 分析可能由 AI 辅助生成，依据公开仓库元数据和 README 的短摘要。 我们不镜像完整 README、文档、Issue 或社媒评论。
cloudflare/computer 被追踪为 TypeScript 项目，主要属于 AI Agent, Infrastructure 方向。这个评估结合公开 GitHub 元数据、分类信号、短来源摘要和 Git-Stars 编辑规则，而不是复制项目文档。
增长检查：该仓库目前有 8.3k Star，今日 +0，本周 +2.0k，本月 +0。这些窗口用于区分持续采用信号和短期曝光峰值。
维护检查：当前活跃度为 一般；最近一次推送距今 61 天，未关闭 Issue 为 14，约占总 Star 的 0.17%。这只是采用信号，不替代工程尽调。
采用检查：450 Fork 和 8.3k Watcher 反映项目被复用和关注的程度。许可证信号：MIT。商业或内部使用前请核验许可证兼容性。
适用判断：当你需要「TypeScript 技术栈团队评估生态原生工具」时，这个项目更值得评估；如果「需要法律审查、安全审计或生产 SLA 保证」，则需要谨慎。
来源检查：Git-Stars 当前为这份报告保留了 2 个明确来源引用，近期增长信号为 2.0k。最终安装、安全和版本信息仍应以原始 GitHub 仓库为准。
原创分析
充分已经具备问题、能力和结论层面的原创判断。
来源
充分已附带仓库元数据和 README/来源引用。
许可证
充分已记录 MIT，可作为评估依据。
维护
充分仓库元数据中可以看到近期活跃信号。
同类对比
充分已有足够相近项目用于对比。
热度
8.3k 星标
复用
450 复刻
关注
8.3k 关注者
维护
moderate
许可证
MIT
未解决 Issue
14
Cloudflare Computer is a virtual filesystem that lives inside a Durable Object, with the authoritative state in SQLite and a pluggable execution surface through `workspace.runtime`. It ships three backends: a container with a FUSE mount, an isolate shell running just-bash, and an isolate JavaScript runtime, all accessible via a single execution entry point. It is a preview package for experimentation, not production use.
核心特性
- Multiple execution backends: container (FUSE mount), isolate shell (just-bash), and isolate JavaScript (ECMAScript modules). - Single execution entry point: `workspace.runtime.exec(source, { backend })`. - Workspace can be used without a backend for filesystem-only operations.
AI Agent
Agent frameworks, autonomous workflows, and tool-use systems
Infrastructure
Cloud, deployment, networking, containers, and platform tooling
npm install @cloudflare/computer63
健康评分
一般
提交活跃度
Jun 5, 2026
创建于
Aug 14, 2026
最近提交
每日同步积累数据后，Star 历史将在此显示。
复刻
未解决
关注者
cloudflare
GitHub 主页freeCodeCamp/freeCodeCamp
freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.
awesome-selfhosted/awesome-selfhosted
A list of Free Software network services and web applications which can be hosted on your own servers
obra/superpowers
An agentic skills framework & software development methodology that works.
react/react
The library for web and native user interfaces.
affaan-m/ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
✓
许可证
✓
Fork 情况
~ 一般
维护情况
解决的问题
It solves the problem of giving AI agents a persistent, stateful filesystem and execution environment that can be accessed across different runtimes. Unlike traditional approaches that require separate storage and compute, it unifies state and execution in a single Durable Object, enabling seamless synchronization and multiple execution backends without additional infrastructure.
能力范围
Developers can build AI agents that perform file operations, run shell commands, execute JavaScript modules, and generate artifacts like PDFs or images. Real-world use cases include automated document generation, code project scaffolding, and interactive chat agents with a working directory. The ceiling is high: it supports full Linux userland in containers, isolated shell environments, and JavaScript execution, all backed by a durable filesystem, enabling complex agent workflows that require persistent state and diverse execution capabilities.
结论
This framework is ideal for developers building AI agents on Cloudflare Workers who need a persistent, synchronized filesystem and flexible execution backends. It is not suitable for production use yet due to its preview status and unstable APIs. The key trade-off is the convenience of a unified state and execution model versus the current lack of production readiness and potential complexity of managing multiple backends.