MCP servers · least privilege

MCP servers with no destructive tools

3,151 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. 1,622 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 30 Oct 2026. https://modelsatlas.com/mcp/no-destructive/page/79

Verdicts · 3,151 servers

✓SAFE1,634!CAUTION522✕BLOCK995

Transports

A server can list more than one transport.

3,151 servers · 1,873–1,896 · ranked by trust score, then stars
Uniprof!CAUTION B

indragiek/uniprof

Universal CPU profiler designed for humans and AI agents

★ 407GitHub stars1 writestdio
Opik!CAUTION B

comet-ml/opik-mcp

Model Context Protocol (MCP) server for Opik, the open-source LLM observability and evaluation platform, built by Comet. Read traces, log scores, and manage prompts from Claude Code, Cursor, or VS Code.

★ 219GitHub stars2 readstdio
opik-mcp!CAUTION B

comet-ml/opik-mcp

Model Context Protocol (MCP) server for Opik, the open-source LLM observability and evaluation platform, built by Comet. Read traces, log scores, and manage prompts from Claude Code, Cursor, or VS Code.

★ 219GitHub stars2 readstdio
BrowserStack!CAUTION B

browserstack/mcp-server

BrowserStack's Official MCP Server

★ 151GitHub stars12 writestdio
ZeroSlop!CAUTION B

manavmishra/ZeroSlop

Open-source Agent Skill that finds and removes AI slop from your writing

★ 132GitHub stars2 readstreamable-http
Auth0!CAUTION B

auth0/auth0-mcp-server

Connect LLMs and AI agents to Auth0 Management APIs via an MCP server. Use natural-language commands to create apps, deploy Actions, and query logs—runs locally and supports interactive tenant authentication for secure Auth0 administration from AI tools/CLIs.

★ 121GitHub stars10 writestdio
Next!CAUTION B

vertile-ai/next-mcp-server

Help LLMs to understand your Next apps better

★ 99GitHub stars4 readsse
Next.js Router Analyzer!CAUTION B

jazelly/next-mcp-server

Help LLMs to understand your Next apps better

★ 99GitHub stars4 readsse
Mindpilot!CAUTION B

abrinsmead/mindpilot-mcp

See through your agent's eyes. Visualize legacy code, architect new systems, understand everything.

★ 90GitHub stars2 readstdio
Multi-AI Advisor!CAUTION B

yuchenssr/multi-ai-advisor-mcp

council of models for decision

★ 88GitHub stars2 readstdio
SO-ARM100 Robot Control!CAUTION B

ilialarchenko/robot_mcp

A simple MCP server for the SO-ARM100 control

★ 85GitHub stars2 read
Google Calendar!CAUTION B

takumi0706/google-calendar-mcp

Google Calendar MCP server for Claude Desktop integration

★ 59GitHub stars4 write
SAP Note Search!CAUTION B

marianfoo/mcp-sap-notes

MCP Server to search and retrieve SAP Notes

★ 57GitHub stars2 readstdio
TomTom!CAUTION B

tomtom-international/tomtom-mcp

A Model Context Protocol (MCP) server providing TomTom's location services, search, routing, and traffic data to AI agents.

★ 56GitHub stars14 readstdio
Uptime Kuma!CAUTION B

davidfuchs/mcp-uptime-kuma

A Model Context Protocol (MCP) server for Uptime Kuma version 2.

★ 54GitHub stars31 readstdio
One MCP!CAUTION B

particlefuture/1mcpserver

★ 52GitHub stars5 readstdio
Instagram Server Next✓SAFE B

duhlink/instagram-server-next-mcp

★ 51GitHub stars1 writestdio
GitGuardian!CAUTION B

gitguardian/ggmcp

MCP server for remediating hardcoded secrets using GitGuardian’s API. It detects over 600 secret types and prevents credential leaks before code is made public.

★ 36GitHub stars2 writestdio
GitGuardian!CAUTION B

gitguardian/gg-mcp

MCP server for remediating hardcoded secrets using GitGuardian’s API. It detects over 600 secret types and prevents credential leaks before code is made public.

★ 36GitHub stars2 writestdio
TrueNAS Core✓SAFE B

vespo92/truenascoremcp

A working TrueNAS Core MCP Server

★ 32GitHub stars1 read
BeatOS!CAUTION B

averatec0773/beatos

Local-first beat library built for producers — catalog, BPM/key analysis, license tiers & PDFs, tagged-MP3 + loopkit export, and assisted multi-platform publishing. Chat with your catalog via a built-in AI Agent (Claude · ChatGPT · DeepSeek), or wire it to any MCP client. Desktop + web.

★ 29GitHub stars5 writestdio
Chrome Google Search!CAUTION B

cmann50/mcp-chrome-google-search

★ 25GitHub stars2 readstdio
Gemini Client!CAUTION B

angrysky56/gemini-mcp-client

A MCP (Model Context Protocol) client that uses Google Gemini AI models for intelligent tool usage and conversation handling. Tested working nicely with Claude Desktop as an MCP Server currently. Based on untested AI gen code by a non-coder use at own risk.

★ 22GitHub stars2 writestdio
hono-mcp-server✓SAFE B

ninjacazul/hono-mcp-server

🔧 Expose Hono API endpoints as MCP tools, simplifying integration and enhancing your API's functionality.

★ 2GitHub stars1 writestreamable-http

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 (1,622 servers). This list also includes servers whose tools write: 1,529 of the 3,151 do.
Does no destructive tools mean safe?
No. 1,634 are SAFE, 522 CAUTION and 995 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.