展示 HN：ctx – 一种代理式开发环境 (ADE)

展示 HN：ctx – 一种代理式开发环境 (ADE)
Show HN: ctx – an Agentic Development Environment (ADE)

## ctx：一种代理开发环境 ctx 是一个统一平台，适用于使用多个 AI 编码代理的团队，例如 Claude Code、Codex 和 Cursor。它为工程师提供单一界面，同时为安全和平台团队提供受控的、容器化的运行时环境，具有强大的隔离和审查能力。主要优势包括：简化对各种代理的访问、在隔离容器中安全执行以及集中管理记录、差异和任务历史。ctx 支持本地和远程开发环境，允许用户使用自己的提供商和凭据，而无需为标准工作流程创建帐户。通过标准化 ctx，团队可以在不分散工作流程的情况下利用首选代理，保持一致的安全控制，并简化审查流程。它设计易于采用——从一个小任务开始，以验证完整的流程。

## ctx：一种新的代理开发环境 Luca-ctx 推出 **ctx (ctx.rs)**，一种专注于代理工作流的新型开发环境——本质上是利用人工智能代理来辅助编码。它采用多线程、基于工作区的界面，并带有容器化工作空间和远程主机模型，允许多代理协作，并具有本地合并队列。早期讨论强调了一个关于 **GitHub Copilot 订阅支持** 的关键问题，原因是 API 成本。另一位用户将其与现有的 ADE（代理开发环境）如 Conductor 进行比较，指出 **代码导航和强大的 IDE 类似体验（如 IntelliJ）** 对于成功至关重要。创建者澄清说，ctx 旨在融合强大 IDE 和多任务代理能力的优势。该工具专为开发者设计，不同于更面向最终用户的解决方案，例如自托管 Replit 的替代品。

原文

ctxis an Agentic Development Environment (ADE)

Any coding agent. Containerized workspaces. Unified transcripts and review.

ctx is an Agentic Development Environment (ADE) for teams using multiple coding agents. It gives engineers one interface for the real Claude Code, Codex, Cursor, and more. It gives security and platform teams one controlled runtime with containerized disk and network isolation, one review surface, and durable transcripts.

Use ctx on your own machine or against a remote devbox or VPS you control. For normal local workflows, you do not need a ctx account, and you can bring your own providers, models, and credentials.

What ctx helps you do

Use Claude Code, Codex, Cursor, and other coding agents in one interface.
Run agents in isolated containers with explicit disk and network controls.
Let agents work with bounded autonomy instead of constant approval prompts.
Keep tasks, sessions, diffs, transcripts, and artifacts in one review surface.
Run work locally or on remote machines you control.
Keep parallel tasks isolated in separate worktrees and land them cleanly with the agent merge queue.

Why standardize on the environment

Engineers can use the agents they prefer without fragmenting the workflow.
Security and platform teams can rely on one runtime model and one set of safety controls.
Review, provenance, and task history stay in one place instead of being scattered across tools.
You can change harnesses and models over time without rebuilding team workflows around each one.

Get started

Use one small, low-risk task to validate the whole loop end to end: install, connect a provider, open a workspace, run a task, and review the diff before you finalize changes.

Good first tasks

update one label or validation message
fix one obvious bug in a narrow area
make one small UI, docs, or config change