Smithery

Overview

Smithery is the go-to registry for MCP (Model Context Protocol) servers — think of it as npm for AI agent tools. Instead of hunting down MCP server configurations, writing JSON by hand, or cloning random repos, you install what you need with a single command and your agent starts using it immediately.

What It Does

Your AI agent can only act on the tools you give it. MCP servers are the bridge between your agent and external services — databases, APIs, file systems, browsers. Smithery turns the process of finding and connecting these tools into a frictionless CLI experience.

The key features:

• One-command install — npx @smithery/cli install server-name downloads, configures, and connects the server
• Registry with 1000+ servers — from web scraping (Firecrawl, Puppeteer) to databases (PostgreSQL, SQLite) to APIs (GitHub, Slack, Gmail)
• Automatic configuration — no manual JSON editing; Smithery writes your MCP config and handles auth
• Version management — update, rollback, and remove servers cleanly
• Compatible with any MCP host — Claude Code, Cursor, Cline, VS Code extensions, and custom agents

How to Use It

# Install the CLI
npx @smithery/cli --help

# Browse available servers interactively
npx @smithery/cli list

# Install a server (e.g., PostgreSQL)
npx @smithery/cli install postgres

# Remove a server
npx @smithery/cli remove postgres

# Update all installed servers
npx @smithery/cli update

Once installed, your agent detects the new MCP server automatically. Ask it to “query the database” or “search the web” and it knows which tool to reach for.

Who Needs This

Anyone running MCP-compatible AI agents who wants to extend their capabilities without manual configuration. If you’ve ever spent 20 minutes wrestling with a claude_desktop_config.json file just to give your agent web search access, Smithery saves you that time every single time.

Bottom Line

Smithery is the package manager your AI agent infrastructure needs. With over 45,000 installs and a growing registry of 1,000+ MCP servers, it’s become the standard way to connect agents to the tools they need. Install it once, and your agent’s capabilities grow on demand.

For a deeper look at how MCP servers work alongside agent-to-agent protocols, check out our A2A Protocol vs MCP comparison.