Skip to content

Build on REM CloudHub

Everything in the console is also in the REST API and the MCP server. Script your deploys, read logs from CI, or let an AI agent operate your projects with a preview first.

REST API

JSON over https, under /api.

Send your token as a bearer token. Every error has the same shape, {"detail": ...}, and validation errors name the field.

curl https://remcloudhub.com/api/apps \
  -H "Authorization: Bearer $RCH_TOKEN"

Tokens

  • Personal tokens (qchp_) act as you, with your permissions. Create them in the console under Team, then Access tokens.
  • Service tokens (qchsa_) belong to a service account, for products and automations; they can be bound to areas, an app and origins.
  • Short and revocable. Give every token an expiry; revoke it from the same page and it stops working at once.

MCP server for AI agents

Point an MCP client at https://remcloudhub.com/mcp with a personal token. The agent sees only the tools its person may call: more than six hundred, from listing projects to reading logs. Tools that change something answer with a preview and a confirmation code that works once, within two minutes. Every call is recorded in Activity.

{
  "mcpServers": {
    "rch": {
      "url": "https://remcloudhub.com/mcp",
      "headers": { "Authorization": "Bearer <your qchp_ token>" }
    }
  }
}

AI gateway

Your apps call a model name such as rch-fast or rch-advanced with an AI key from the console's AI page, in the OpenAI chat-completions format. The gateway tries the provider keys behind the name in order; usage is paid from your organisation's REM credits and shown per key, model and day.

Get a token and try it

Sign in, create a personal token, and call the API or connect your agent.