面向AI AGENT

你的AI Agent已经知道如何使用的基础设施。

CLI优先。基于SSH。默认JSON输出。每个rdc命令都为自主Agent设计——无需GUI、无需浏览器自动化、无需自定义集成。

免费开始使用阅读文档 →

$ rdc machine status prod-1 -o json

✓ SSH connection established .... done

✓ Querying renet daemon ......... done

✓ 12 repositories, 47 containers .. all healthy

✓ {"success": true, "command": "machine info", "data": {...}}

Format: JSON Latency: 142ms Auth: SSH Tokens: ~200

输出仅作示例，实际运行可能有更多日志。 CLI reference: rdc machine

100%

CLI-native operations

~200

Avg. tokens per response

Browser automation needed

问题所在

AI Agent功能强大——但大多数基础设施在与之对抗

你的AI编程助手可以自主编写、测试和调试代码。但当它需要部署、检查状态或管理基础设施时，就会遇到障碍：仅限GUI的仪表板、OAuth浏览器流程、不稳定的API，以及撑爆上下文窗口的响应。

73% of developers plan to use AI agents for infrastructure tasks by 2026 GitHub 2025 ^[1]

4x more context window used by unstructured vs. JSON output Industry analysis ^[2]

0% of traditional infrastructure platforms designed for AI agents Market survey ^[3]

传统基础设施

Step 1 Open dashboard

Step 2 OAuth login

Step 3 Parse HTML

Step 4 Custom API

Step 5 Agent fails

使用 Rediacc

rdc repo up

JSON

Done

工作原理

三步实现Agent管理的基础设施

Configure

Drop an AGENTS.md template into your project root. Your AI assistant reads it on startup and knows every rdc command.

Discover

Run rdc agent capabilities to get a structured JSON command tree. The agent knows what's available without documentation.

Operate

Deploy, monitor, rollback — all through CLI commands with JSON output. No GUI, no browser, no custom integration.

AI Agent Running

claude-code CLI

rdc repo up SSH

json-output stdout

agent-caps API

4 个命令 ~200 tokens

rdc

SSH + JSON

Infrastructure Managed

prod-1 12 repos

staging 8 repos

backups snapshots

monitoring health

47 个容器 2.4 TB

原生AGENT设计

为什么rdc是为AI Agent而生

Rediacc的每一个设计决策都优先考虑机器可读接口。非TTY检测自动切换为JSON输出。一致的信封格式意味着Agent以相同方式解析每个响应。基于SSH的认证意味着无需OAuth流程或令牌管理。

传统平台

Rediacc (rdc CLI)

GUI dashboards: requires browser automation or screenshots

CLI-first: every operation is a shell command

OAuth/browser login: agents can't authenticate

SSH keys: agents authenticate like any CLI tool

HTML/unstructured output: 4x token usage, fragile parsing

JSON envelope: consistent schema, ~200 tokens per response

No discovery: agents need docs or hardcoded API calls

rdc agent capabilities: structured command tree at runtime

为何重要

为自主基础设施管理而构建

Zero integration effort

Drop AGENTS.md in your project. Your AI assistant immediately knows every available command, flag, and workflow.

Context-window friendly

JSON responses average ~200 tokens. The --fields flag lets agents request only the data they need. No wasted context.

Safe by default

--dry-run previews destructive operations. --yes enables automation. Agents can deploy confidently with built-in guardrails.

给你的一条消息

让你的AI Agent管理你的基础设施

安装CLI。添加AGENTS.md模板。你的AI助手会处理其余一切。

免费开始使用

$ curl -fsSL https://www.rediacc.com/install.sh | bash