MCP servers · connect in Claude Code

Claude Code MCP servers

4,770 MCP servers have a connect command for Claude Code. 2,136 are SAFE, 937 CAUTION and 1,697 BLOCK. Every one was read in full before it was listed.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Claude Code MCP servers”. Retrieved 30 Oct 2026. https://modelsatlas.com/mcp/for/claude-code/page/138

Verdicts · 4,770 servers

✓SAFE2,136!CAUTION937✕BLOCK1,697

Transports

A server can list more than one transport.

Tool risk

1,345

servers expose at least one destructive tool: delete, overwrite or run. The median server with a readable tool list exposes 10 tools.

Servers with no destructive tools →
4,770 servers for Claude Code · 3,289–3,312 · ranked by trust score, then stars
X Social✕BLOCK D

infatoshi/x-mcp

MCP server for the X (Twitter) API -- give AI agents the ability to post, search, read, and engage on X

★ 54GitHub stars2 destructivestdio
Mermaid✕BLOCK D

narasimhaponnada/mermaid-mcp

★ 53GitHub stars1 writestdio
Sitecore Server✕BLOCK D

antonytm/mcp-sitecore-server

Model Context Protocol server for Sitecore

★ 53GitHub stars13 destructivestdio
A11y✕BLOCK D

priyankark/a11y-mcp

An MCP (Model Context Protocol) server for performing accessibility audits on webpages using axe-core. Use the results in an agentic loop with your favorite AI assistants (Amp/Cline/Cursor/GH Copilot) and let them fix a11y issues for you!

★ 52GitHub stars2 readstdio
Chart MinIO✕BLOCK D

zaizaizhao/mcp-server-chart-minio

mcp-server-chart私有化部署方案

★ 52GitHub stars1 destructive
Dotbot✕BLOCK D

andresharpe/dotbot

Structured, auditable AI-assisted development for teams. Zero-dependency MCP server, web dashboard, and multi-provider AI CLI support.

★ 52GitHub stars5 read
Immich Photo Manager✕BLOCK D

drolosoft/immich-photo-manager

MCP server and Claude Code plugin for self-hosted Immich: CLIP and OCR search, geographic albums, duplicates, people and faces, metadata repair, PDF photobooks. 94 tools, tested live on Immich 2.x and 3.x.

★ 52GitHub stars13 destructivestdio
Matrix✕BLOCK D

mjknowles/matrix-mcp-server

MCP Server for a Matrix home server integration; chat, manage rooms, etc.

★ 52GitHub stars5 writestreamable-http
Playwright Cdp✕BLOCK D

lars-hagen/mcp-playwright-cdp

Model Context Protocol server for Playwright with Chrome DevTools Protocol support

★ 52GitHub stars1 destructivestdio
Perplexity Comet✕BLOCK D

rapiercraft/perplexity-comet-mcp

MCP server connecting Claude Code to Perplexity Comet browser. Autonomous web browsing, agentic research, multi-tab management. Windows/WSL/macOS support. Smart completion detection, auto-reconnect, tab protection. Enhanced fork with 7 tools including comet_tabs.

★ 51GitHub stars3 writestdio
Binance✕BLOCK D

analyticace/binance-mcp-server

Unofficial tools and server implementation for Binance's Model Context Protocol (MCP). Designed to support developers building crypto trading AI Agents.

★ 50GitHub stars2 writestdio
Binance✕BLOCK D

analyticace/binancemcpserver

Unofficial tools and server implementation for Binance's Model Context Protocol (MCP). Designed to support developers building crypto trading AI Agents.

★ 50GitHub stars2 writestdio
Widget✕BLOCK D

ref-tools/widget-mcp

Simple MCP-UI widgets for common use cases.

★ 50GitHub stars4 readstdio
@putervision/state-memory-mcp✕BLOCK D

putervision/state-memory-mcp

Persistent, branch-aware workflow state memory MCP server for AI coding assistants. Tracks tasks, accepted decisions, and active blockers to prevent session context bloat and speed up development.

★ 49GitHub stars1 readstdio
Agentmap✕BLOCK D

raymondchins/agentmap

Stop your coding agent reading the wrong files. Compiler-grade TS/JS repo map — 100% precision on blast radius vs grep's 60%, measured on public repos. CLI + MCP server, fully local, no vector DB.

★ 49GitHub stars1 writestdio
RLM✕BLOCK D

richardwhiteii/rlm

Recursive Language Model patterns for Claude Code — handle massive contexts (10M+ tokens) by treating them as external variables

★ 49GitHub stars1 writestdio
Widemem✕BLOCK D

remete618/widemem-ai

Local-first memory layer for LLM apps and agents. Runs offline on Ollama, sentence-transformers and FAISS; OpenAI and Anthropic optional. Importance scoring, contradiction resolution, confidence-aware retrieval.

★ 49GitHub stars1 destructivestdio
ZenML✕BLOCK D

zenml-io/mcp-zenml

MCP server to connect an MCP client (Cursor, Claude Desktop etc) with your ZenML MLOps and LLMOps pipelines

★ 49GitHub stars1 readstdio
Pagecast✕BLOCK D

mcpware/pagecast

Record any browser page as GIF or video via MCP — powered by Playwright + ffmpeg

★ 48GitHub stars3 writestdio
Puppeteer Vision✕BLOCK D

djannot/puppeteer-vision-mcp

★ 48GitHub stars1 readsse
Engram✕BLOCK D

tstockham96/engram

Universal memory layer for AI agents. Remember, recall, consolidate.

★ 47GitHub stars5 writestdio
McpDocServer✕BLOCK D

ruan11223344/mcpdocserver

一个基于MCP协议的开发文档服务器,专为各类开发框架文档设计

★ 47GitHub stars3 readstdio
Motif✕BLOCK D

jvogan/motif

An AI-native molecular biology workbench for Claude Science: Claude opens sequence files and prepares analyses on request; you annotate, align, and review them in an interactive workspace.

★ 47GitHub stars2 writestdio
Rosetta✕BLOCK D

tikoci/rosetta

MCP Server with RouterOS docs + commands + products + changelogs, using SQLite-as-RAG, sourced from MikroTik

★ 47GitHub stars16 readstdio

More about Claude Code

Other client hubs

By risk

Questions

How do I add an MCP server to Claude Code?
Run the claude mcp add command shown on the server's page; it names the package and its version.
Which Claude Code MCP servers are safe?
Of the 4,770 servers with a Claude Code connect command, 2,136 are SAFE, 937 CAUTION and 1,697 BLOCK. Cards are ranked by trust score, so the safest are first; open one to read its findings before you connect it.
Are any of these servers remote?
1,942 of the 4,770 list a remote transport (streamable HTTP or SSE). The rest run as a local process your client starts.

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.