Cloud World Model
模擬雲端基礎設施，並在部署或變更基礎設施前估算成本、延遲與韌性。AI代理可比較AWS、GCP、Azure、OCI及DigitalOcean上的架構，無需配置真實資源。
託管 MCP 伺服器
npx add-mcp 'https://www.cloudworldmodel.ai/mcp'可安裝到 Claude Code、Codex、Cursor 等客戶端
文件
MCP Server: Simulate Cloud Infrastructure and Estimate Cost
Cloud World Model is a hosted streamable HTTP MCP server that lets coding agents simulate cloud infrastructure and estimate cost, latency, errors and resilience before deploying or changing infrastructure. It covers AWS, GCP, Azure, OCI and DigitalOcean. Simulations don't create resources in your cloud account.
Without an API key, 9 keyless demo tools are available. To unlock all 63 authenticated tools, add the x-api-key header.
{"mcpServers":{"cloud-world-model":{"url":"https://www.cloudworldmodel.ai/mcp","headers":{"x-api-key":"<your CWM API key>"}}}}
Connect via Smithery
Search for 'Cloud World Model' on Smithery and click Connect. Your AI client will be configured automatically with no manual setup.
Paste the hosted endpoint URL
Alternatively, paste the hosted HTTP MCP URL directly into any MCP-compatible client (Cursor, Windsurf, Claude Desktop, or Continue).
Configure in Cursor or Windsurf
Add the MCP server URL to your Cursor or Windsurf settings under MCP Servers. Restart the IDE to activate the Cloud World Model tools.
Try it with an API key
Register a World Model key on the Get Started page to unlock full API access and remove the anonymous demo rate limit.
Full agent example
Working end-to-end agent example that simulates an infrastructure change before deployment in two parts: a small REST simulation paid per call with x402 on Base ($0.003, verified with a real paid run), and an advanced keyless MCP walkthrough showing a Redis cache failure before/after.