MCP servers · connect in Claude Desktop

Claude Desktop MCP servers

3,432 MCP servers have a connect command for Claude Desktop. 1,660 are SAFE, 638 CAUTION and 1,134 BLOCK. Every one was read in full before it was listed.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Claude Desktop MCP servers”. Retrieved 29 Oct 2026. https://modelsatlas.com/mcp/for/claude-desktop/page/141

Verdicts · 3,432 servers

✓SAFE1,660!CAUTION638✕BLOCK1,134

Transports

A server can list more than one transport.

Tool risk

1,009

servers expose at least one destructive tool: delete, overwrite or run. The median server with a readable tool list exposes 9 tools.

Servers with no destructive tools →

3,432 servers for Claude Desktop

3,361–3,384 of 3,432 · ranked by trust score, then stars

who2be

luetzey/who2be

✕BLOCK F

Self-hosted hub for AI agent configuration: versioned personas, playbooks and knowledge with a mandatory review step, served to agents over MCP. Agent configuration you review — not behavior you hope for.

★ 05 readstdio

Genkit

genkit-ai/genkit

✕BLOCK F

Open-source framework for building agentic apps in JavaScript, Go, Dart, and Python, built and used in production by Google

★ 6,5091 destructivesse

SoloMD

zhitongblog/solomd

✕BLOCK F

A markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.

★ 1,130Not listed in sourcestdio

AGI Memory

cognitivecomputations/agi-memory

✕BLOCK F
★ 6173 readstdio

caura

caura-ai/caura

✕BLOCK F

Caura (formerly MemClaw) — governed shared memory for AI agent fleets. Multi-agent, multi-tenant, MCP-native. Trust tiers, keystone policies, audit trails, knowledge graph, self-improving retrieval. Apache 2.0.

★ 5442 writestreamable-http

One MCP

1mcp-app/agent

✕BLOCK F

A unified Model Context Protocol server implementation that aggregates multiple MCP servers into one.

★ 5093 destructivesse

jazz-ai

lvndry/jazz

✕BLOCK F

One agent. Every surface. Your rules.

★ 541 destructivestreamable-http

claude-autopm

rafeekpro/ClaudeAutoPM

✕BLOCK F

Autonomous Project Management Framework for Claude Code - Advanced AI-powered development automation

★ 395 writestdio

unitares

cirwel/unitares

✕BLOCK F

Accountability infrastructure for long-running AI agents. Self-hosted MCP server: identity, claims and evidence, review, outcomes, and reconstruction across agent runtimes.

★ 55 readstdio

viboplr

outcast1000/viboplr

✕BLOCK F

Plugin-first desktop music player for macOS and Windows — local library, streaming services, lyrics, and radio. Built with Tauri 2, React, and Rust.

★ 31 write

N8n

czlonkowski/n8n-mcp

✕BLOCK F

A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you

★ 22,9493 destructivesse

App Store Connect

joshuarileydev/app-store-connect-mcp-server

✕BLOCK F
★ 3311 destructivestdio

jev-judge-mcp

PyModel/jev-judge-mcp

✕BLOCK F

Typed judgment tools for MCP agents. TypeSafe's Jev model as verify, screen, find, classify, rerank, decide, compare, extract, review, gate, and score: the model judges, policy decides auto, review, or escalate.

★ 711 readstdio

Digital Twins

pome-sh/digital-twins

✕BLOCK F

Test mode for your integrations, built for the way agents build

★ 211 destructivestreamable-http

agent-bounties

NSPG13/agent-bounties

✕BLOCK F

Open-source bounty network where AI agents post, fund, claim, solve, verify, and get paid for digital work.

★ 185 writestreamable-http

Theokit Sdk

usetheokit/theokit-sdk

✕BLOCK F

Open-source TypeScript SDK and agent runtime you own end to end. Agent.create / prompt / stream / resume, 43 LLM providers on your own keys, MCP with OAuth PKCE, subagents, workflows, evals, memory, and a bubblewrap+seccomp kernel sandbox for tool execution. Local-first, opt-in cloud, Apache-2.0.

★ 313 writestdio

KXM

kontextmind/kxm

✕BLOCK F

Pi extensions for multi-agent communication and orchestration

★ 22 writestdio

n.360

resper1965/n.360

✕BLOCK F

n.iso — plataforma de adequação ISO 27001/27701 da ness.

★ 011 writestdio

Genkit

firebase/genkit

✕BLOCK F

Open-source framework for building agentic apps in JavaScript, Go, Dart, and Python, built and used in production by Google

★ 6,4521 destructivesse

ReverseLab

ling71671/open-reverselab

✕BLOCK F

Open-source AI reverse-engineering agent platform and MCP server for Ghidra, Frida, x64dbg and Rizin — automated PE/APK/binary analysis, CTF and malware research, with 100+ MCP tools and a 194-article runnable knowledge base.

★ 1,1826 destructive

open-reverselab

LING71671/open-reverselab

✕BLOCK F

Open-source AI reverse-engineering agent platform and MCP server for Ghidra, Frida, x64dbg and Rizin — automated PE/APK/binary analysis, CTF and malware research, with 100+ MCP tools and a 194-article runnable knowledge base.

★ 1,1486 destructive

Circuit

snowfort-ai/circuit-mcp

✕BLOCK F

Computer use-like MCP for webapps and electron apps, to enable AI agents to test their changes

★ 607 writesse

AFD — Agent-First Development

lushly-dev/afd

✕BLOCK F
★ 416 destructivestdio

kernl

fastslack/kernl

✕BLOCK F

Self-hosted agent teams that do your work — and hand any LLM your real data (tasks, mail, calendar, finance) over MCP. No account, no cloud, nothing leaves your machine.

★ 336 destructivestdio

More about Claude Desktop

Other client hubs

By risk

Questions

How do I add an MCP server to Claude Desktop?
Paste the JSON block shown on the server's page into the mcpServers section of Claude Desktop's configuration file.
Which Claude Desktop MCP servers are safe?
Of the 3,432 servers with a Claude Desktop connect command, 1,660 are SAFE, 638 CAUTION and 1,134 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?
1,261 of the 3,432 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.