MCP Servers Directory

Model Context Protocol (MCP) servers connect AI clients to external tools and data. Compare each server's published tools, connection transport, authentication requirements, installation notes, and source before connecting it.

MCP Servers

Browse all MCP servers

3 servers

Playwright MCP

Automate and inspect web pages through Playwright’s browser MCP tools.

Transport: stdio (default); optional local HTTP mode

Last reviewed: 2026-10-10

GitHub MCP Server

Give AI agents scoped access to GitHub repositories, issues and pull requests.

Transport: stdio (local) and HTTP (hosted)

Last reviewed: 2026-10-10

Context7 MCP Server

Fetch current library and framework documentation inside AI coding tools.

Transport: stdio (local) and Streamable HTTP (hosted)

Last reviewed: 2026-10-10

How AI Dots reviews entries

We check the primary repository or official documentation, record the supported clients and setup requirements, and revisit version-sensitive details. Popularity labels are used only when the public signal and its date are recorded.

What to compare

Check the server's declared tools, transport, authentication, required credentials, data access, source repository, and whether setup instructions match the current version.

Sources to explore

Use the official MCP Registry and protocol documentation to verify server metadata and connection details. Public directories are discovery aids; check the server's own source before granting access.

Frequently asked questions

What is an MCP server?

An MCP server exposes tools or data through the Model Context Protocol so a compatible AI client can connect to them.

What should I check before connecting an MCP server?

Review its source, declared tools, transport, authentication, credentials, and data access. Only connect servers you trust and understand.

Does AI Dots verify every server release?

Each profile shows its last review date. Server behavior can change between releases, so check the linked primary source before connecting.