MCP server

CloudHike.dev runs an MCP (Model Context Protocol) server. Once you add it to your AI editor, the agent gets CloudHike.dev tools it can call by itself: find or create a project, deploy Postgres, Auth, Storage and Redis, wait until they are ready, put the .env into your app, look at your tables and files, and read these docs. No prompt to paste and no curl.

1. Open Connect MCP in the dashboard sidebar (https://cloudhike.dev/dashboard/coding-agent).
2. Pick your editor and click Copy with key. This creates your workspace API key the first time.
3. Paste the snippet where the card says (a terminal command or a config file), then restart or reload your editor. Ask it: "Use the cloudhike tools to set up the backend for this app."

Replace icu_your_api_key with your key, or use Copy with key in the dashboard, which fills it in.

Claude Code

claude mcp add --transport http --scope user cloudhike https://cloudhike.dev/api/mcp \
  --header "Authorization: Bearer icu_your_api_key"

Then run /mcp inside Claude Code to check it is connected.

Codex

# 1. Add this line to your shell profile (~/.zshrc or ~/.bashrc):
#    export CLOUDHIKE_API_KEY="icu_your_api_key"
# 2. Add this to ~/.codex/config.toml:
[mcp_servers.cloudhike]
url = "https://cloudhike.dev/api/mcp"
bearer_token_env_var = "CLOUDHIKE_API_KEY"

Open a new terminal so the variable is set, then run codex mcp list to check it.

Cursor

{
  "mcpServers": {
    "cloudhike": {
      "url": "https://cloudhike.dev/api/mcp",
      "headers": {
        "Authorization": "Bearer icu_your_api_key"
      }
    }
  }
}

Open Cursor Settings → MCP to check it shows a green dot.

VS Code

{
  "inputs": [
    {
      "type": "promptString",
      "id": "cloudhike-key",
      "description": "CloudHike.dev API key",
      "password": true
    }
  ],
  "servers": {
    "cloudhike": {
      "type": "http",
      "url": "https://cloudhike.dev/api/mcp",
      "headers": {
        "Authorization": "Bearer ${input:cloudhike-key}"
      }
    }
  }
}

VS Code asks for the API key the first time and keeps it in its secret storage. Use Copilot Chat in Agent mode.

Windsurf

{
  "mcpServers": {
    "cloudhike": {
      "serverUrl": "https://cloudhike.dev/api/mcp",
      "headers": {
        "Authorization": "Bearer icu_your_api_key"
      }
    }
  }
}

Gemini CLI

{
  "mcpServers": {
    "cloudhike": {
      "httpUrl": "https://cloudhike.dev/api/mcp",
      "headers": {
        "Authorization": "Bearer icu_your_api_key"
      }
    }
  }
}

Claude Desktop

{
  "mcpServers": {
    "cloudhike": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://cloudhike.dev/api/mcp",
        "--header",
        "Authorization:${AUTH_HEADER}"
      ],
      "env": {
        "AUTH_HEADER": "Bearer icu_your_api_key"
      }
    }
  }
}

Needs Node.js. Restart Claude Desktop after saving.

Any other MCP client

Point it at https://cloudhike.dev/api/mcp with the header Authorization: Bearer <key>. Clients that only run local (stdio) servers can use npx -y mcp-remote https://cloudhike.dev/api/mcp --header "Authorization:Bearer <key>", as in the Claude Desktop example.

The server also offers each docs page as a resource (cloudhike://docs/<page>) and a setup_backend prompt. In editors that show MCP prompts as slash commands, run it to set up the whole backend in one go.

• On every plan, query_database runs in a READ ONLY transaction and accepts one statement at a time, so an agent cannot change data or tables with it.
• On the Pro plan, the agent also sees execute_sql, which can insert, update, delete and change the schema. The whole script runs in one transaction that is only committed if every statement succeeds. Editors ask you before running it, because it is marked as destructive.
• Every execute_sql run is written to the audit log as postgres.sql-write.
• Database tools share a limit of 60 calls per 10 minutes.

• The key can create projects, deploy services and read keys. Nothing can be deleted through MCP: deleting projects and services stays in the dashboard.
• get_environment returns secrets so the agent can write your env file. The server tells agents to write them only to the git-ignored env file and never to repeat them in chat. Every fetch is written to the audit log as environment.agent-fetch.
• All tools follow your plan limits and rate limits, exactly like the dashboard.
• Keep the key out of files you commit. The VS Code example asks for the key and stores it in VS Code's secret storage, so .vscode/mcp.json is safe to commit. The other examples go in your home folder.

To test the server yourself, run npx @modelcontextprotocol/inspector, choose Streamable HTTP, enter the server URL and add the Authorization header.