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/111

Verdicts · 3,145 servers

✓SAFE1,632!CAUTION521✕BLOCK992

Transports

A server can list more than one transport.

3,145 servers

2,641–2,664 of 3,145 · ranked by trust score, then stars

aicode-toolkit

AgiFlow/aicode-toolkit

✕BLOCK D

Toolkit for Coding Agents to work reliably with repo of any size.

★ 1622 writesse

AI Code Toolkit

agiflow/aicode-toolkit

✕BLOCK D

Toolkit for Coding Agents to work reliably with repo of any size.

★ 1612 writesse

Mnemo Cortex

guymanndude/mnemo-cortex

!CAUTION D

Open-source cognitive coprocessor with active memory for AI agents — persistent recall, semantic search, overnight dreaming, verified facts, encrypted USB sync. MCP server; works with Claude, ChatGPT, and any local LLM. Built by one maker and his agents.

★ 1574 writestdio

Universal Tool Calling Protocol (UTCP) for TypeScript

universal-tool-calling-protocol/typescript-utcp

✕BLOCK F

Official typescript implementation of UTCP. UTCP is an open standard that lets AI agents call any API directly, without extra middleware.

★ 1193 writestdio

U.S. Census Bureau Data API

uscensusbureau/us-census-bureau-data-api-mcp

✕BLOCK D

The U.S. Census Bureau Data API MCP connects AI Assistants with official Census Bureau statistics.

★ 9811 readstdio

AI Agents Orchestrator

hoangsonww/ai-agents-orchestrator

✕BLOCK D

Intelligent orchestration system that coordinates multiple AI coding assistants (Claude, Codex, Gemini CLI, Copilot CLI) to collaborate on complex software development tasks via REPL or a Vue/Nuxt UI dashboard. Also includes an Agentic Team runtime with role-based multi-agent open communication & le

★ 9012 read

Poke Gate

f/poke-gate

✕BLOCK D

Expose your machine to your Poke AI assistant via MCP tunnel

★ 794 write

Ctxo

alperhankendi/ctxo

✕BLOCK F

MCP server for AI coding agents. Instead of reading files one by one, your agent gets dependency graphs, git intent, blast radius, and change health in a single call. Works with any language deep analysis for TypeScript,Java, Go, and C#.

★ 6914 readstdio

1xn AI Workflow Composer

1xn-labs/1xn-vmcp

✕BLOCK F

vMCP - Virtual Model Context Protocol

★ 5111 writestdio

Mcp Logic

angrysky56/mcp-logic

✕BLOCK D

Fully functional AI Logic Calculator utilizing Prover9/Mace4 via Python based Model Context Protocol (MCP-Server)- tool for Windows, Linux, Claude App etc

★ 464 readstdio

Prompt Decorators

synaptiai/prompt-decorators

✕BLOCK F

A standardized framework for enhancing how LLMs process and respond to prompts through composable decorators, featuring an official open standard specification and Python reference implementation. Claude Code plugin and MCP server integration.

★ 442 write

Congress

amurshak/congressmcp

!CAUTION F

An MCP server allowing AI agents and MCP clients to interface with the Congress.gov API

★ 3324 readstreamable-http

Sourced

ryancodrai/sourced

✕BLOCK F

Maps packages to source code. Allows coding agents to search any dependency with grep.

★ 321 writestreamable-http

SSH Orchestrator

samerfarida/mcp-ssh-orchestrator

✕BLOCK F

Secure SSH access for AI agents via MCP. Execute commands across your server fleet with policy enforcement, network controls, and comprehensive audit logging.

★ 283 writestdio

desktop-touch-mcp

Harusame64/desktop-touch-mcp

✕BLOCK F

Windows computer-use MCP server: drive any desktop app via semantic discover-then-act targeting (entities + leases, not pixel coordinates), with per-action perception guards, a native Rust UIA engine, Chrome CDP, and Key Locker credential autofill for ssh/sudo password prompts. Works with Claude, Cu

★ 221 writestdio

A2A ADK

rubenszimbres/a2a_adk_mcp

✕BLOCK F

Multi-Agent Systems with Google's Agent Development Kit + A2A + MCP

★ 22 writestdio

openswissdata

cammac-creator/openswissdata

✕BLOCK F

Swiss federal reference data for AI agents: customs tariff (TARES), FINMA register and warnings list, NOGA/NACE/ISIC crosswalks. Free MCP server, no key. Signed dataset files.

★ 04 readstdio

Cloudflare

cloudflare/mcp-server-cloudflare

✕BLOCK F
★ 4,2602 writestreamable-http

iai Personal Memory Engine

codeabra/iai-personal-memory-engine

!CAUTION F

A cyber brain for your AI. It never forgets a detail, remembers exactly what you said, and learns how you work over time. Free, local, works with Cursor, Claude Code, Codex, OpenClaw, Hermes and more. MIT.

★ 8962 writestdio

ssh-mcp

tufantunc/ssh-mcp

✕BLOCK F

MCP server exposing SSH control for Linux servers via Model Context Protocol.

★ 7993 writestdio

bagel

Extelligence-ai/bagel

✕BLOCK F

Query robotics, drone, and IoT data in plain English through an MCP server, with an intelligent edge data reduction pipeline that keeps only the data that matters.

★ 3971 readsse

HomeButler

higangssh/homebutler

✕BLOCK F

🏠 Tells you what changed on your server — only what's worth telling. Single Go binary, no database, nothing running on the machines it watches. MCP server built in.

★ 2981 readstdio

ClawMem

yoloshii/clawmem

✕BLOCK F

On-device memory layer for AI agents. Claude Code, OpenClaw and Hermes. Hooks + MCP server + hybrid RAG search.

★ 2113 writestdio

Jules SDK

google-labs-code/jules-sdk

✕BLOCK F
★ 13812 writestdio

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.