cloudish Skill
在 Cloudish 上部署和运行应用——这是一朵 AI 智能体会部署的云。它会在服务端构建 Dockerfile、源代码文件夹或容器镜像,并提供一个可访问的线上 URL,支持数据库和文件的持久化存储卷、加密密钥管理,以及通过 Cloudish 的 OIDC 提供方为你的应用用户实现登录。智能体无需注册即可自行生成 API key,人类用户可以稍后认领。当用户要求在 Cloudish 上部署、托管或上线应用、网站、API、机器人或数据库服务,或获取 Cloudish API key 时使用。
安装方式:把技能目录放入 ~/.claude/skills/(Claude Code)或在 claude.ai 设置中启用;也可复制右侧安装命令一键添加。
技能指令原文(SKILL.md)
Deploy to Cloudish
Cloudish is the cloud an AI agent deploys a container to. The agent creates its
own API key with one unauthenticated call, Cloudish builds the image server-side
(no local Docker needed), and runs it as a long-lived container with an optional
persistent volume, paid from prepaid credits the key can never exceed.
Source of truth
The full, current instructions — endpoints, request fields, auth, pricing — are
at https://cloudish.ai/skill.md. Fetch and follow it before making any Cloudish
API call. Do not guess endpoints or fields; if it can't be fetched, say so and
stop rather than improvise.
Workflow
- Inspect the project: entrypoint, port, Dockerfile, env vars, data that must persist.
- Prefer an existing Dockerfile; otherwise write a minimal one for the stack.
- Make sure the server binds to
0.0.0.0on its declared port. - Reuse
CLOUDISH_API_KEYfrom.envif present; if you mint one, save it to a
gitignored .env before anything else.
- Never print keys or secrets into chat, source, or logs.
- Confirm before anything that spends credits beyond what the user already asked for.
- Deploy per https://cloudish.ai/skill.md, then report the live URL and anything
that matters about persistence, env vars, or networking.
- If a deploy fails, read the build/runtime logs and fix the cause; never claim
success without seeing it.