Knowledge base
MCPCloud is an MCP server factory and skill studio. Paste an API spec, and Claude enriches every tool description while MCPCloud generates a typed MCP server and deploys it to a global edge runtime — no scaffolding, no handoff. This knowledge base explains the concepts behind that workflow and how teams operate it day to day.
From an imported API spec to a live, agent-ready MCP server in one pass.
Start here
What is MCP?
Model Context Protocol fundamentals: clients, servers, tools, transports.
What MCPCloud provides
The managed platform on top of MCP: factory, studio, registry, runtime.
Key concepts
Projects, servers, deployments, skills, installations, and credits.
Lifecycle
Import → enrich → generate → deploy → operate → evolve.
How it works
Server factory
The pipeline that turns an OpenAPI, GraphQL, or repo spec into a deployed server.
Skill Studio
Compose reusable, versioned agent behaviors with triggers and guardrails.
Runtime access
Public, privateOrg, and unlisted deployments plus runtime token exchange.
Transports & auth
Streamable HTTP, SSE, stdio, and built-in auth and static analysis.
Marketplace & registry
Publish, discover, version, and monetize servers and skills.
Credits & plans
How credits, limits, and plan tiers work.
Operational support
Troubleshooting
Fast diagnostics for build, deploy, auth, and CLI issues.
FAQ
Common questions about the platform, CLI, and runtime model.