MCP servers · least privilege

MCP servers with no destructive tools

3,145 MCP servers declare at least one tool and none of them is classed destructive: nothing deletes, overwrites or runs a command through a declared tool. 1,620 of them only read; the rest may create or edit.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “MCP servers with no destructive tools”. Retrieved 29 Oct 2026. https://modelsatlas.com/mcp/no-destructive/page/29

Verdicts · 3,145 servers

✓SAFE1,632!CAUTION521✕BLOCK992

Transports

A server can list more than one transport.

3,145 servers

673–696 of 3,145 · ranked by trust score, then stars

Wireshark

khuynh22/mcp-wireshark

✓SAFE B

An MCP server that integrates Wireshark/tshark with AI tools and IDEs. Capture live traffic, parse .pcap files, apply display filters, follow streams, and export JSON - all via Claude Desktop, VS Code, or CLI. Cross‐platform, typed, tested, and pip‐installable.

★ 631 writestdio

Aula

casperjuel/aula-mcp

✓SAFE B

MCP server for Denmark's Aula school platform — TypeScript MitID auth, no headless browser. Exposes profiles, calendar, messages, ugeplaner to AI agents (Claude/Cursor/etc) via Model Context Protocol.

★ 621 writesse

Baba Is Eval

lennart-finke/baba_is_eval

✓SAFE B

Claude et al. play the brilliant puzzle title "Baba is You"

★ 622 write

Browser Use

jovanipink/mcp-browser-use

✓SAFE B

FastAPI MCP server that exposes browser-use automation through a testable Model Context Protocol boundary.

★ 621 write

Devto

arindam200/devto-mcp

✓SAFE B

MCP Server of DevTo

★ 622 write

Magento 2

boldcommerce/magento2-mcp

!CAUTION B

A MCP server for Magento 2

★ 622 writestdio

NS Travel Information

r-huijts/ns-mcp-server

✓SAFE B

A Model Context Protocol (MCP) server that provides access to NS (Dutch Railways) travel information through Claude AI. This server enables Claude to fetch real-time train travel information and disruptions using the official Dutch NS API.

★ 626 readstdio

RAG App on AWS

genieincodebottle/rag-app-on-aws

✓SAFE B

Build and deploy a full-stack RAG app on AWS with Terraform, using free tier Gemini Pro, real-time web search using Remote MCP server and Streamlit UI with token based authentication.

★ 622 readstreamable-http

Shell Command

egoist/shell-command-mcp

✓SAFE B

MCP server for executing shell commands.

★ 621 writestdio

Voice Call

popcornspace/voice-call-mcp-server

✓SAFE B
★ 621 writestdio

AI Humanizer

text2go/ai-humanizer-mcp-server

✓SAFE B

A powerful Model Context Protocol (MCP) server that helps refine AI-generated content to sound more natural and human-like. Built with advanced AI detection and text enhancement capabilities.

★ 611 readstdio

Apple Books

vgnshiyer/apple-books-mcp

✓SAFE B

Apple Books MCP Server

★ 619 readstdio

Astro

morinokami/astro-mcp

✓SAFE B

MCP server to support Astro project development

★ 611 writesse

Cline Memory Bank

dazeb/cline-mcp-memory-bank

✓SAFE B

A memory system for Cline that tracks progress between conversations.

★ 613 writestdio

Cryptocurrency Market Data

nayshins/mcp-server-ccxt

✓SAFE B

Cryptocurrency Market Data MCP Server

★ 617 readstdio

FFmpeg Video Processor

bitscorp-mcp/mcp-ffmpeg

✓SAFE B
★ 614 readstdio

Hub

warpdev/mcp-hub-mcp

✓SAFE B
★ 617 readstdio

Jimengpic

comeonzhj/jimengpic-mcp

✓SAFE B

使用即梦 API 生成图片的 MCP 服务

★ 611 readstdio

LLM Wiki Kit

iamsashank09/llm-wiki-kit

✓SAFE B

Stop re-explaining your research to your AI agent. Persistent, LLM-maintained wikis that compound over time. Drop PDFs, URLs, YouTube - your agent remembers forever. Based on Karpathy's LLM Wiki pattern.

★ 611 writestdio

Screenshot

upnorthmedia/screenshotmcp

✓SAFE B

A Model Context Protocol MCP server for capturing website screenshots with full page, element, and device size features.

★ 613 readstdio

Agentloop

luckeyfaraday/athena-loops

✓SAFE B

MIT-licensed, backend-agnostic AI agent orchestration loop in Python: orchestrator→worker→reviewer as a deterministic harness. Drive coding-agent backends through an MCP server + CLI.

★ 605 write

BatchIt

ryanjoachim/mcp-batchit

✓SAFE B

🚀 MCP aggregator for batching multiple tool calls into a single request. Reduces overhead, saves tokens, and simplifies complex operations in AI agent workflows.

★ 601 writestdio

Domain Availability Checker

imprvhub/mcp-domain-availability

✓SAFE B

A Model Context Protocol (MCP) server that enables Claude Desktop to check domain availability across 50+ TLDs. Features DNS/WHOIS verification, bulk checking, and smart suggestions. Zero-clone installation via uvx.

★ 602 readstreamable-http

HMR

promplate/hmr

✓SAFE B

Real hot-module reload for Python—side effects handled reactively. https://py3.online/hmr

★ 601 readsse

Other risk hubs

By job

Questions

What counts as destructive?
A tool that deletes, overwrites or runs commands. We class every tool the source declares as read, write or destructive; a server is listed here when it has tools and none is destructive.
How is this different from read-only?
Read-only servers have only read tools (1,620 servers). This list also includes servers whose tools write: 1,525 of the 3,145 do.
Does no destructive tools mean safe?
No. 1,632 are SAFE, 521 CAUTION and 992 BLOCK on this list. The verdict comes from the audit of the source, not from the tool list alone.

Keep going in your AI

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