Credal
The control plane for governed enterprise AI agents.
The Docker Hub of MCP: discover, install, and host thousands of MCP servers.
Smithery is arguably the closest thing to a Docker Hub for MCP, and its combination of a big registry plus one-click hosted endpoints is genuinely useful for skipping local setup. The CLI and OAuth handling for hosted servers are nice touches. Caveats: server quality varies widely across the long tail, hosted execution costs apply beyond browsing, and the registry landscape (Smithery, Glama, PulseMCP, the official registry) is still shaking out. Verify hosting limits and pricing with the vendor.
Smithery is a registry and hosting platform for MCP servers, letting developers search thousands of servers, install them via CLI, or run them as hosted remote endpoints. It manages runtime and OAuth for hosted servers and offers a Toolbox meta-MCP to route agents. Browsing and publishing are free, with paid tiers for hosted execution.
Smithery is one of the largest MCP registries, indexing thousands of servers (6,000+ and growing) that developers can search, install locally through its CLI, or run on Smithery's own infrastructure as hosted, streamable-HTTP remote endpoints. For hosted servers it manages the runtime and provides OAuth modals so server authors don't build auth flows from scratch and consumers don't install dependencies. Beyond discovery, Smithery aims to be the distribution layer for MCP, offering deployment tooling (hosted and external HTTP, plus stdio) and a Toolbox meta-MCP that routes agents to appropriate servers. Browsing and publishing are free; hosted execution and higher usage move to paid tiers. As with all MCP infrastructure in 2026, the registry and hosting details evolve quickly.
Smithery is a large MCP registry plus hosting layer, letting you search thousands of servers and run them as hosted endpoints with managed OAuth, free to browse but paid for hosted execution.
Smithery operates a central hub for the MCP ecosystem, combining discovery, a CLI, and managed hosting. It aims to be the standard distribution channel for MCP servers.
The platform serves both server authors who want distribution and consumers who want tools without operating infrastructure.
Core features include a searchable registry of thousands of servers, CLI-based local installation, and hosted remote endpoints over streamable HTTP with managed runtime and OAuth.
A Toolbox meta-MCP routes agents to appropriate servers, and deployment tooling supports hosted, external HTTP, and stdio server types.
MCP developers and AI application builders who need to find, distribute, or run MCP servers, plus server authors seeking distribution.
Developers browsing and running MCP servers.
Engineering teams standardizing on MCP tooling.
MCP community, server authors, and AI IDE users.
A developer or small team building on MCP who wants a fast path to discover and run servers without managing local infrastructure.
Smithery is an early-stage venture; specific funding details should be verified as public information may be incomplete or outdated.
A registry and hosting platform for MCP servers, letting you discover, install, and run thousands of servers, including as hosted remote endpoints.
No. You can install via the CLI or use Smithery's hosted option to get a remote streamable-HTTP endpoint with no local dependencies.
Yes, browsing and publishing are free. Hosted execution and higher usage move to paid tiers.
Any MCP-aware client, including Claude Desktop, Cursor, Windsurf, and VS Code, can consume Smithery-hosted endpoints.
A router that helps agents discover and select the right server among many, reducing manual configuration.
Side-by-side pages for pricing, features, and best-fit use cases.
The control plane for governed enterprise AI agents.
Claude is an AI chatbot developed by Anthropic, offering conversational AI capabilities with multiple model tiers (Opus, Sonnet, Haiku) and specialized products including Claude Code for software deve
ChatGPT is an AI chatbot developed by OpenAI that handles conversational tasks including writing, coding, learning, and brainstorming
AI research assistant to find, understand, and write about papers