Deploy from the chat
The agent finds or creates a project and deploys Postgres, Auth, Storage, Redis, Jobs or Email, with the same plan and billing checks as the dashboard.
Add the CloudHike.dev MCP server to your AI editor. The agent can create a project, deploy services, write your .env, look at your tables and files, and read the docs. No prompt to paste and no curl.
Connect Claude Code, Codex, Cursor, VS Code and other AI editors to your backend with MCP.
The agent finds or creates a project and deploys Postgres, Auth, Storage, Redis, Jobs or Email, with the same plan and billing checks as the dashboard.
get_environment gives the agent your keys, with instructions to write them only to the git-ignored env file and not repeat them in chat. Every fetch is in the audit log.
List tables, describe columns and run read-only SQL, up to 500 rows. On Pro, execute_sql can change data in one transaction, after you approve it.
get_docs returns any docs page as Markdown, so the agent writes code from the real API. A setup_backend prompt does the whole setup in one go.
Snippets for Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI and Claude Desktop. Any client that speaks Streamable HTTP works.
Nothing can be deleted through MCP, and every tool follows your plan and rate limits. Revoke the key to disconnect every editor at once.
Open Connect MCP in the dashboard, pick your editor and click Copy with key. The first copy creates your workspace API key.
Run the command or add it to your editor's config file, then restart or reload the editor.
Say "Use the cloudhike tools to set up the backend for this app." It deploys services and writes the code that connects them.
claude mcp add --transport http --scope user cloudhike https://cloudhike.dev/api/mcp \
--header "Authorization: Bearer icu_your_api_key"
# Then ask Claude Code:
# "Use the cloudhike tools to set up the backend for this app."Same project, same .env, same bill.
Postgres, auth, storage, Redis, email and AI models. One API key.