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/11

Verdicts · 3,145 servers

✓SAFE1,632!CAUTION521✕BLOCK992

Transports

A server can list more than one transport.

3,145 servers

241–264 of 3,145 · ranked by trust score, then stars

Taisly Agent Kit

taisly/agent

✓SAFE B

Taisly Agent Kit: MCP server, CLI, SDK, and agent docs for publishing videos to TikTok, Reels, Shorts, X, and Facebook.

★ 2175 writestdio

Gibber

anton10xr/gibber-mcp

✓SAFE B

Tiny MCP server with cryptography tools, sufficient to establish end-to-end encryption between LLM agents

★ 2164 readsse

Gibber

antonpk1/gibber-mcp

✓SAFE B

Tiny MCP server with cryptography tools, sufficient to establish end-to-end encryption between LLM agents

★ 2164 readsse

Claude Mermaid

veelenga/claude-mermaid

✓SAFE B

MCP Server to previewing mermaid diagrams with live reload

★ 2151 writestdio

Chiasmus

yogthos/chiasmus

!CAUTION B

Chiasmus is an MCP server that gives language models access to formal verification

★ 2144 writestdio

Figma

matthewdailey/figma-mcp

✓SAFE B

ModelContextProtocol for Figma's REST API

★ 2132 writestdio

Code Executor

bazinga012/mcp_code_executor

!CAUTION B

The MCP Code Executor is an MCP server that allows LLMs to execute Python code within a specified Conda environment.

★ 2125 writestdio

Lighthouse

priyankark/lighthouse-mcp

!CAUTION B

Allows AI assistants such as Amp/Cline/Cursor/Claude Code/Codex/GitHub Copilot to use Google's lighthouse tool to measure perf metrics for your webpage. You can then run an agentic loop and get the assistants to optimize those metrics!

★ 2101 writestdio

OpenNutrition

deadletterq/mcp-opennutrition

✓SAFE B

MCP server providing access to the comprehensive OpenNutrition food database with 300,000+ food items, nutritional data, and barcode lookups

★ 2074 readstdio

Magic UI

magicuidesign/mcp

✓SAFE B

Magic UI components, searchable and installable from your AI editor.

★ 2063 readstdio

BaseLayer

zwgnr/baselayer

✓SAFE B

Re-usable multi part components built on React Aria and TailwindCSS.

★ 2051 read

Code Sandbox

philschmid/code-sandbox-mcp

✓SAFE B
★ 2042 writestdio

Mcp Rb

funwarioisii/mcp-rb

✓SAFE B

A lightweight Ruby framework for building MCP servers with a Sinatra-like DSL

★ 2043 read

HotNews

wopal-cn/mcp-hotnews-server

✓SAFE B

A Model Context Protocol server that provides real-time hot trending topics from major Chinese social platforms and news sites.

★ 20110 readstdio

Itasca

yusong652/itasca-mcp

✓SAFE B

MCP server connecting AI agents to ITASCA engines (PFC, FLAC, 3DEC, MPoint, MassFlow) — run geotechnical & geomechanics simulations through natural conversation

★ 2012 readstdio

PFC

yusong652/pfc-mcp

✓SAFE B

MCP server connecting AI agents to ITASCA engines (PFC, FLAC, 3DEC, MPoint, MassFlow) — run geotechnical & geomechanics simulations through natural conversation

★ 2012 readstdio

Simple Arxiv

andybrandt/mcp-simple-arxiv

✓SAFE B

Tool to work with arXiv, provide LLM with ability to search and read papers from there

★ 2011 writestreamable-http

TikTok

seym0n/tiktok-mcp

✓SAFE B

Model Context Protocol (MCP) with TikTok integration

★ 2011 writestdio

GCP

eniayomi/gcp-mcp

✓SAFE B

A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with your Google Cloud Platform environment. This allows for natural language querying and management of your GCP resources during conversations.

★ 2001 writestdio

gcp-mcp

eniayomi/gcp-mcp

✓SAFE B

A Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with your Google Cloud Platform environment. This allows for natural language querying and management of your GCP resources during conversations.

★ 2001 writestdio

Abap ADT

mario-andreschak/mcp-abap-adt

✓SAFE B

ABAP ADT MCP server for reading SAP objects (programs, classes, tables, CDS) via ADT

★ 1961 writestdio

Claude Patent Creator

robthepcguy/claude-patent-creator

!CAUTION B

USPTO patent creation system with MCP server + Claude Code plugin. Hybrid RAG search over MPEP/USC/CFR, BigQuery access to 76M+ patents, automated 35 USC 112 compliance checks, prior art search, diagram generation. GPU-accelerated with skills and autonomous agents.

★ 1969 read

Easy

zcaceres/easy-mcp

✓SAFE B

Absurdly easy Model Context Protocol Servers in Typescript

★ 1967 readstdio

use-webmcp-tool

GoogleChromeLabs/use-webmcp-tool

✓SAFE B
★ 1951 write

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.