Skip to content
Stork.AIstork.ai
AI Tool

Smithery Review

Smithery is a marketplace for Model Context Protocol (MCP) servers, enabling AI agents to connect to external services and extend their capabilities.

shipped Apr 2, 2026aifreemium
ai
Smithery — product screenshot

Why it matters

1Smithery hosts over 2,880 MCP servers, facilitating integration with external services.
2The platform supports major integrations including GitHub, Gmail, PostgreSQL/SQL, and Brave Search.
3A security vulnerability in October 2025 compromised over 3,000 MCP servers, exposing API keys.
4Smithery offers a freemium pricing model with specific rate limits for basic MCP server usage.

Stork’s verdict on Smithery

Smithery offers a standardized plug-and-play system for AI agents, but hosted servers suffered a critical API key exposure in October 2025.

About Smithery

Business Model
Subscription SaaS
Platforms
Web, CLI
Target Audience
Developers and businesses looking to enhance AI capabilities
API DocsGitHub

Specs

API Docs

View Documentation →

API Available

Yes, public API

overview

What is Smithery?

Smithery is an AI tool developed by Smithery (company) that enables AI agent developers to discover, deploy, and manage Model Context Protocol (MCP) compliant services. It serves as a centralized hub for integrating language model agents with external tools and data sources, standardizing tool configurations and interfaces for AI applications. The platform functions as a registry and management system for MCP servers, which are modular components designed to extend the capabilities of large language models (LLMs) by connecting them to various external services and data sources. Smithery aims to simplify the development and deployment of agentic AI applications by providing a standardized way for AI agents to interact with the 'outside world', cataloging MCP servers and offering optional cloud hosting for these services.

features

Key Features of Smithery

Smithery provides a comprehensive set of features designed to facilitate the development and deployment of AI agents using the Model Context Protocol (MCP). These features streamline the process of connecting AI agents to external services and managing their interactions.

  • Access to over 2,880 MCP servers for extending AI agent capabilities.
  • Management of OAuth and session tokens for secure external service connections.
  • Command Line Interface (CLI) for easy integration and local development workflows.
  • Pre-built workflows to define and automate agent behavior.
  • Observability tools for monitoring the performance and usage of published MCP tools.
  • Optional cloud hosting for MCP servers, making them broadly accessible.
  • Support for uploading and hosting 'Skills' directly on the platform, simplifying deployment.
  • Standardized tool configurations and interfaces for consistent AI application development.

use cases

Who Should Use Smithery?

Smithery is primarily designed for developers and businesses focused on building and enhancing AI agent capabilities, particularly those leveraging the Model Context Protocol (MCP).

  • AI Agent Developers: For discovering and deploying MCP-compliant agentic services, and integrating language model agents with external tools and data sources like GitHub, Gmail, and databases.
  • Developers Creating Agentic Services: For hosting and distributing custom MCP servers, and standardizing tool configurations and interfaces for AI applications.
  • Businesses Enhancing AI Capabilities: For extending AI agents with new capabilities and skills, and automating workflows using MCPs to manage tasks and schedules with various integrations.

how to use

How to Use Smithery

Getting started with Smithery involves leveraging its platform to discover, deploy, and manage MCP servers, either locally or via cloud hosting. The process typically begins with exploring the available MCP tools and integrating them into an AI agent's workflow.

  • 1Explore MCP Servers: Browse the Smithery marketplace to discover over 2,880 available MCP-compliant agentic services.
  • 2Install CLI: Utilize the Smithery Command Line Interface (CLI) for local development and integration of MCP servers.
  • 3Integrate with Agents: Connect selected MCP servers to your AI agents to extend their capabilities with external tools and data sources.
  • 4Manage Authentication: Configure OAuth and session management for secure access to integrated services.
  • 5Deploy Skills: Upload and host custom 'Skills' directly on the Smithery platform for simplified deployment and distribution.
  • 6Monitor Usage: Use the platform's observability features to track the performance and interactions of your deployed MCP tools.

pricing

Smithery Pricing & Plans

Smithery operates on a freemium business model, offering free access for basic usage with specific rate limits, alongside paid tiers for more extensive features and higher usage. While specific rate limits for the general Smithery Connect API are not publicly detailed, individual MCPs may have their own usage-based pricing, such as 'rate-limit-manager' at $0.001 per request. The platform also offers 'Pro' and 'Custom' pricing tiers, which provide enhanced capabilities and support for advanced use cases, though specific pricing for these tiers is not publicly disclosed.

  • Freemium: Free usage for basic MCP servers with rate limits.
  • Pro: Paid tier with enhanced features and higher usage limits (specific pricing not detailed).
  • Custom: Tailored plans for enterprise and specific requirements (specific pricing not detailed).

Pros

  • +Provides a centralized registry and easy installation mechanism for thousands of MCP servers, simplifying AI agent integration.
  • +The open-source nature of its CLI and TypeScript SDK allows for developer inspection and contribution.
  • +Simplifies AI development by enabling quick integration of MCP tools that connect LLMs to external services.
  • +Offers a well-regarded security model for local installations, keeping tokens on the user's machine.
  • +Seen as a 'missing piece in the AI agent puzzle' due to its standardized, modular plug-and-play system for AI agents.
  • +Supports a rapid ecosystem growth, with over 7,600 extensions and support for major platforms like Cursor, VS Code, and Claude Desktop.

Cons

  • −A critical security vulnerability in October 2025 compromised over 3,000 hosted MCP servers, exposing API keys.
  • −Early versions of the Smithery CLI faced community concerns regarding minified code, hindering auditing.
  • −Lack of clarity regarding exact token handling or refresh processes for hosted MCP servers, with recommendations for local secret management.
  • −Some competitor observations suggest a growth strategy involving non-functional Dockerfiles submitted to GitHub, potentially for awareness rather than utility.

Similar Tools

Smithery vs Competitors

Smithery positions itself as a central hub for Model Context Protocol (MCP) servers, aiming to standardize how AI agents connect to external services. Its primary differentiator is its explicit focus on the MCP as a universal standard for AI tool integration and its marketplace for MCP-compliant services.

1
ComposioOn StorkCompare

Composio is a developer-first platform providing a unified framework with over 1000 pre-built connectors and managed authentication for building production-grade AI agents that connect to external tools and services.

Composio offers a comprehensive, developer-focused platform with a large number of pre-built integrations and managed MCP servers, directly competing with Smithery's marketplace for MCP servers by providing a ready-to-use ecosystem for agent tool calling. It operates on a freemium model, similar to Smithery.

2
ArcadeOn StorkCompare

Arcade is specifically focused on AI agent tool calling using the Model Context Protocol (MCP), offering a pre-built catalog of tools and an SDK for building custom tools and custom MCP servers.

Arcade is a very direct competitor to Smithery due to its explicit focus on MCP and providing a catalog of tools, whereas Smithery is described as a marketplace for MCP servers. Its pricing model is described as complex and based on 'user challenges' and 'tool executions,' which differs from a straightforward freemium structure.

3
Zapier (Natural Language Actions)On StorkCompare

Zapier provides access to over 9,000 apps and 20,000 actions through a natural language API interface and pre-built MCP servers, simplifying AI agent integration with a vast ecosystem.

Zapier offers a significantly larger ecosystem of pre-built integrations and MCP servers compared to Smithery, targeting a broader audience of both developers and non-technical users for AI agent automation. It also has a freemium model for its core automation platform, with AI agent capabilities integrated into this offering.

4
LangChain / LangSmithOn StorkCompare

LangChain provides an open-source framework for building AI agents, while LangSmith offers an engineering platform for observing, evaluating, and deploying them, including support for extending with MCP servers.

While LangChain is primarily a framework, its extensive tool ecosystem and LangSmith's platform capabilities, including MCP server integration, make it a strong competitor for developers building and deploying AI agents that connect to external services, similar to Smithery's offering but with a broader development and operational focus. LangChain is open-source, and LangSmith offers commercial services.

Visit Smithery↗
Connect
𝕏
X / Twitterx.com/SmitheryDotAi
⌘
GitHubgithub.com/smithery-ai
💬
Discorddiscord.gg/sKd9uycgH9

AI Reputation Report

Is Smithery yours?

ChatGPT, Perplexity, Gemini, Claude & Grok answer buyer questions about Smithery every day. See whether they name Smithery — or send buyers to a rival.

See what AI saysfree preview

Featured in articles