Cloudwright

Please install the `cloudwright` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install the Cloudwright MCP package** — Install the published MCP server package from PyPI. Run: `pip install cloudwright-ai-mcp` (https://pypi.org/project/cloudwright-ai/)
- **Set an LLM provider API key** — Required for LLM-backed Cloudwright features such as design, modify, chat, and adr. Cloudwright supports Anthropic or OpenAI.

Canonical MCP server config (stdio transport):
- command: `cloudwright`
- args: ["mcp"]
- optional environment variables:
  - `ANTHROPIC_API_KEY`: Anthropic API key for Anthropic-backed LLM features. (example: `<your-anthropic-api-key>`)
  - `OPENAI_API_KEY`: OpenAI API key for OpenAI-backed LLM features. (example: `<your-openai-api-key>`)
  - `CLOUDWRIGHT_LLM_PROVIDER`: Optional LLM provider override. Set to choose the backend explicitly, for example openai; otherwise Cloudwright can auto-detect from available API keys. (example: `<llm-provider>`)
  - `CLOUDWRIGHT_MODEL`: Optional model override for the selected LLM provider. (example: `<model-name>`)

Note: The README explicitly documents starting the MCP server with `cloudwright mcp` and labels it as stdio. At least one LLM API key is practically required for the default design/chat-style MCP functionality, though some non-LLM Cloudwright commands work offline outside MCP usage.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.