Model Context Protocol
MCP Servers
MCP servers are how AI agents connect to tools, data, and capabilities. The public registry grew to 9,400+ servers in 2026, with 78% of enterprise AI teams running at least one in production. Browse the best, including AI coding skills via the skills-hub MCP server.
Short answer
MCP servers expose tools and data to AI clients (Claude Code, Cursor, Codex, Windsurf) through a standard JSON-RPC interface. The skills-hub MCP server is the simplest way to add AI coding skills to any MCP-compatible tool, run `npx @skills-hub-ai/mcp` or point your client at the remote endpoint at https://api.skills-hub.ai/mcp.
Top skills via the skills-hub MCP server
01code-review
661 installsThorough code review — checks correctness, security, performance, readability, and test coverage. Gives actionable feedback ranked by severity.
Review02ui-design-system
357 installsUI design system toolkit for Senior UI Designer including design token generation, component documentation, responsive design calculations, and developer handoff tools. Use when creating design systems, generating design tokens, maintaining visual consistency, or facilitating design-dev collaboration and developer handoff.
Buildfrom Multi-Domain Skills03senior-frontend
298 installsFrontend development skill for React, Next.js, TypeScript, and Tailwind CSS applications. Use when building React components, optimizing Next.js performance, analyzing bundle sizes, scaffolding frontend projects, implementing accessibility, or reviewing frontend code quality.
Buildfrom Multi-Domain Skills04frontend-design
214 installsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.
Buildfrom Anthropic05web-research-agent
161 installsA comprehensive autonomous web research agent that performs targeted searches, fetches and parses web content, summarizes findings, and generates structured reports, adhering to industry best practices for accuracy, security, and robustness.
Analysis06Skill Finder
148 installsWorkflow orchestrator. Decomposes a task into ordered steps, finds or installs the right skill for each step, runs the chain, and offers to save it as a reusable skill.
Combo07markitdown
145 installsConverts heterogeneous documents and selected URIs to Markdown with Microsoft MarkItDown for text analysis, search, and LLM/RAG ingestion. Covers safe local conversion, streams, Office/PDF/data formats, batch workflows, plugins, vision OCR, Azure extraction, and the official MCP server.
Researchfrom Scientific Skills08senior-backend
137 installsDesigns and implements backend systems including REST APIs, microservices, database architectures, authentication flows, and security hardening. Use when the user asks to "design REST APIs", "optimize database queries", "implement authentication", "build microservices", "review backend code", "set up GraphQL", "handle database migrations", or "load test APIs". Covers Node.js/Express/Fastify development, PostgreSQL optimization, API security, and backend architecture patterns.
Buildfrom Multi-Domain Skills09security-review
122 installsSecurity audit and vulnerability assessment for any codebase. Scans for authentication bypasses, missing auth middleware, broken JWT validation (algorithm confusion, weak secrets, missing expiry), OAuth state and PKCE flaws, IDOR and horizontal privilege escalation.
Reviewfrom Skills Hub10self-improving-agent
121 installsCurate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern from notes to enforced rules, (3) turning a debugging solution into a skill, (4) checking memory health and capacity.
Buildfrom Multi-Domain Skills11refactor
112 installsSystematic code refactoring — extract functions, reduce complexity, eliminate duplication, improve naming, and split large files. Preserves behavior with tests.
Review12using-superpowers
101 installsUse when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions
Buildfrom Superpowers
Frequently asked questions about MCP servers
What are MCP servers?
MCP servers are Model Context Protocol servers, long-running processes that expose tools, resources, and prompts to AI clients (Claude Code, Cursor, Codex CLI, Windsurf, Cline, etc.) through a standard JSON-RPC interface. Use them to wire AI into databases, APIs, file systems, browser automation, vector memory, and SaaS tools.
How many MCP servers exist in 2026?
The public MCP registry grew from 1,200 in early 2025 to 9,400+ servers in 2026, with 78% of enterprise AI teams running at least one MCP server in production. Big growth areas: edge MCP (Cloudflare), vector-memory MCP (Qdrant), marketing automation MCP (HubSpot, Salesforce, Linear), and Amazon Bedrock AgentCore.
What's the best MCP server for AI coding?
Skills-hub.ai's MCP server (@skills-hub-ai/mcp) exposes AI coding skills as MCP prompts. Run claude mcp add skills-hub -- npx @skills-hub-ai/mcp (Claude Code) or point any MCP client at https://api.skills-hub.ai/mcp (remote).
How do I install an MCP server?
Most MCP servers are distributed via npm or Smithery. For Claude Code: claude mcp add <name> -- npx <package>. For Cursor: edit .cursor/mcp.json. For Windsurf, Cline, GitHub Copilot Chat: their respective MCP configs. The skills-hub MCP server is also available as a remote HTTP endpoint requiring no local install.
Are MCP servers the same as agent skills?
No. MCP servers expose tools and data (read DB, hit API, control browser). Agent Skills (SKILL.md) are reusable natural-language instructions. They compose, the skills-hub MCP server exposes installed skills as MCP prompts, bringing both standards together.
Which AI tools support MCP in 2026?
Claude Code, Claude Desktop, Cursor, Windsurf, Codex CLI, Cline, Continue.dev, GitHub Copilot Chat (via MCP integration), Google Antigravity 2.0 (launched May 19 2026), and AWS Kiro (via Amazon Bedrock AgentCore). 30+ clients total.