Model Context ProtocolThe USB-C for AI
MCP is an open protocol that standardizes how applications provide context to LLMs. Connect AI models to any data source or tool with a single, unified interface.
What is the Model Context Protocol?
Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect devices to peripherals, MCP provides a standardized way to connect AI models to data sources and tools.
Access a growing ecosystem of pre-built integrations that your LLM can directly plug into.
Switch between LLM providers and vendors without rebuilding your integrations.
Best practices for securing your data within your own infrastructure.
MCP Host
Claude Desktop, IDEs, AI Tools
MCP Server A
Database
MCP Server B
File System
MCP Server C
Web APIs
Core MCP Concepts
Understanding the building blocks of the Model Context Protocol
Expose data and content from your servers to LLMs. Resources can be files, database records, API responses, or any structured data.
Create reusable prompt templates and workflows that can be shared across different AI applications.
Enable LLMs to perform actions through your server. Tools can execute commands, make API calls, or interact with external systems.
Let your servers request completions from LLMs, enabling sophisticated multi-step workflows and agent behaviors.
Popular MCP Servers
Discover the most popular MCP servers across different categories
Development Tools
Search repositories, read files, create issues, and manage pull requests directly from your AI assistant.
Integrate with GitLab repositories, manage merge requests, and access project data.
Create and manage Linear issues, track project progress, and sync with your development workflow.
Data Sources
Query PostgreSQL databases, inspect schemas, and analyze data using natural language.
Access SQLite databases, run queries, and explore local data files.
Search, read, and manage Google Drive files and folders from your AI assistant.
Productivity & Communication
Send messages, search conversations, and manage Slack workspaces through AI.
Create pages, search databases, and manage Notion workspaces with AI assistance.
Search notes, create connections, and navigate your Obsidian knowledge base.
System Integration
Read, write, and manage local files and directories through your AI assistant.
Execute commands on remote servers and manage infrastructure through SSH.
Manage Docker containers, images, and compose services with AI assistance.
Zero Configuration
No Playwright setup, no local browsers, no version management
Natural Language Testing
Describe tests in plain English, AI handles the automation
Remote Browsers
Fully managed, secure browsers with tunnel connections
CI/CD Integration
Export tests to run in your deployment pipeline
Example Usage:
"Test the ability to create an account and login"→ AI automatically navigates your app, fills forms, and validates the complete user flow
Ready to Get Started with MCP?
Choose your path based on what you want to build
Build your own MCP server to expose your data and tools to AI models.
Build a ServerIntegrate MCP into your AI application to access all available servers.
Build a ClientStart using pre-built MCP servers in Claude Desktop right away.
Setup Claude Desktop