Best MCP Servers in 2026

Updated February 2026 · Curated from 5,000+ servers

MCP (Model Context Protocol) servers extend AI coding assistants like Claude Code, Cursor, Windsurf, and Cline with external tools -- database access, API integrations, file systems, and more. This is our curated list of the best MCP servers, selected from the 5,000+ servers in our directory.

Contents
  1. Databases & Data
  2. Developer Tools
  3. Cloud & Infrastructure
  4. Search & Knowledge
  5. Communication
  6. File Systems & Storage
  7. AI & Machine Learning
  8. How to Install MCP Servers

Browse the full directory. This list highlights the best MCP servers. For the complete collection, visit the MCP Servers Directory with 5,000+ servers, filterable by category, transport type, and registry.

Databases & Data

PostgreSQL MCP Server

Official · stdio · @modelcontextprotocol/server-postgres

Query PostgreSQL databases directly from your AI assistant. Read schemas, run queries, and explore data without leaving your editor. One of the most widely used MCP servers.

claude mcp add postgres -- npx @modelcontextprotocol/server-postgres postgresql://localhost/mydb

SQLite MCP Server

Official · stdio · @modelcontextprotocol/server-sqlite

Read and write SQLite databases. Great for local development, prototyping, and working with embedded databases in mobile or desktop apps.

Supabase MCP Server

Community · stdio · supabase

Connect to Supabase projects. Query tables, manage auth, and work with storage directly from Claude Code or Cursor.

Developer Tools

GitHub MCP Server

Official · stdio · @modelcontextprotocol/server-github

Interact with GitHub repositories, issues, pull requests, and actions. Essential for any development workflow that touches GitHub.

Linear MCP Server

Community · stdio · linear

Create and manage Linear issues, projects, and cycles. Let your AI assistant handle project management tasks alongside coding.

Sentry MCP Server

Official · stdio · sentry

Pull error reports, stack traces, and performance data from Sentry. Debug production issues faster by giving your AI context about real errors.

Cloud & Infrastructure

AWS MCP Server

Community · stdio · aws

Manage AWS resources -- S3, Lambda, DynamoDB, EC2, and more. Query CloudWatch logs and deploy infrastructure without switching to the AWS console.

Cloudflare MCP Server

Official · stdio · cloudflare

Manage Workers, KV, D1, R2, and other Cloudflare services. Deploy and configure Cloudflare resources from your editor.

Vercel MCP Server

Community · stdio · vercel

Deploy, manage, and monitor Vercel projects. Check deployment status, view logs, and configure domains.

Brave Search MCP Server

Official · stdio · @modelcontextprotocol/server-brave-search

Web search powered by Brave. Give your AI assistant the ability to search the internet for documentation, examples, and solutions.

Puppeteer MCP Server

Official · stdio · @modelcontextprotocol/server-puppeteer

Browser automation and web scraping. Navigate pages, take screenshots, and extract content from web pages.

Communication

Slack MCP Server

Official · stdio · @modelcontextprotocol/server-slack

Read and send Slack messages, search channels, and manage workspace communication from your AI assistant.

Gmail MCP Server

Community · stdio · gmail

Read, search, and draft emails through Gmail. Useful for summarizing threads and drafting responses based on code changes.

File Systems & Storage

Filesystem MCP Server

Official · stdio · @modelcontextprotocol/server-filesystem

Secure file system access with configurable permissions. Read, write, and manage files with sandboxed directory access.

Google Drive MCP Server

Community · stdio · google-drive

Access Google Drive files and folders. Search documents, read spreadsheets, and manage shared drives.

AI & Machine Learning

Qdrant MCP Server

Community · stdio · qdrant

Vector database operations for AI applications. Search embeddings, manage collections, and build RAG pipelines.

LangChain MCP Server

Community · stdio · langchain

Integrate LangChain tools and chains. Build and test AI pipelines with access to LangChain's ecosystem.

How to Install MCP Servers

Installation depends on your AI coding tool:

Claude Code

claude mcp add server-name -- npx @org/mcp-server

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "server-name": {
      "command": "npx",
      "args": ["@org/mcp-server"]
    }
  }
}

VS Code

Add to .vscode/settings.json:

{
  "mcp": {
    "servers": {
      "server-name": {
        "command": "npx",
        "args": ["@org/mcp-server"]
      }
    }
  }
}

For the full installation guide, see our MCP Servers Guide. Browse all servers in the MCP Servers Directory.