MCP servers · least privilege

Read-only MCP servers

1,620 MCP servers declare tools that only read: none of their tools is classed write or destructive. That makes them the safest place to start giving an agent access. Read-only describes the tools, not the code, so the audit verdict still decides whether to connect one.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Read-only MCP servers”. Retrieved 29 Oct 2026. https://modelsatlas.com/mcp/read-only/page/10

Verdicts · 1,620 servers

✓SAFE913!CAUTION261✕BLOCK446

Transports

A server can list more than one transport.

1,620 servers

217–240 of 1,620 · ranked by trust score, then stars

Stitch

kargatharaakash/stitch-mcp

✓SAFE B

Universal MCP Server for Google Stitch. Connect AI agents to your UI designs.

★ 1212 readstdio

FFmpeg

egoist/ffmpeg-mcp

✓SAFE B

An MCP server for FFmpeg

★ 1202 readstdio

Scrapling Fetch

cyberchitta/scrapling-fetch-mcp

✓SAFE B

Helps AI assistants access text content from bot-protected websites. MCP server that fetches HTML/markdown from sites with anti-automation measures using Scrapling.

★ 1202 readstdio

Terraform Registry

thrashr888/terraform-mcp-server

✓SAFE B

Terraform Registry MCP Server

★ 12031 readstdio

Codex Bridge

elyin/codex-bridge

✓SAFE B

A lightweight MCP server bridging AI agents to OpenAI models via official CLI

★ 1183 read

JavaSinkTracer

zacarx/javasinktracer_mcp

✓SAFE B

基于函数级污点分析的 Java 源代码漏洞审计工具JavaSinkTracer,通过 Model Context Protocol (MCP) 为 AI 助手提供安全分析能力。

★ 1186 readstdio

Mureka

skyworkai/mureka-mcp

✓SAFE B

generate lyrics, song and background music(instrumental). Model Context Protocol (MCP) server.

★ 1181 read

deepl-mcp-server

DeepL/deepl-mcp-server

✓SAFE B

A Model Context Protocol (MCP) server that provides translation capabilities using the DeepL API.

★ 11713 readstdio

Posecode

posecode-dev/posecode

!CAUTION B

An open-source text language, parser, validator and Three.js renderer for inspectable 3D human movement.

★ 1173 readstdio

WordPress

emzimmer/server-wp-mcp

✓SAFE B
★ 1171 readstdio

Academic Search

afrise/academic-search-mcp-server

✓SAFE B

Academic Paper Search MCP Server for Claude Desktop integration. Allows Claude to access data from Semantic Scholar and Crossref.

★ 1163 readstdio

DeepL

deeplcom/deepl-mcp-server

✓SAFE B

A Model Context Protocol (MCP) server that provides translation capabilities using the DeepL API.

★ 11613 readstdio

Oura

yuzehao2023/mcp-oura

✓SAFE B

MCP server for Oura API integration

★ 1166 readstdio

Xueqiu

liqiongyu/xueqiu_mcp

✓SAFE B
★ 11645 read

Cli

mcpgod/cli

✓SAFE B

Fine-grained control over model context protocol (MCP) clients, servers, and tools. Context is God.

★ 1151 readstdio

Deepseek Thinking Claude

newideas99/deepseek-thinking-claude-3.5-sonnet-cline-mcp

✓SAFE B

🧠 MCP server implementing RAT (Retrieval Augmented Thinking) - combines DeepSeek's reasoning with GPT-4/Claude/Mistral responses, maintaining conversation context between interactions.

★ 1152 readstdio

Luma

jochenyang/luma-mcp

!CAUTION B

多模型视觉理解 MCP 服务器,为不支持图片理解的 AI 编码模型提供视觉能力:分析截图、报错、UI 与文档,可接入多家主流视觉大模型。Multi-model vision MCP server that adds image understanding to AI coding models without native vision — analyze screenshots, errors, UI and documents via major vision LLM providers.

★ 1151 readstdio

Py-MCP-Naver

pfldy2850/py-mcp-naver

✓SAFE B

python MCP NAVER

★ 11514 read

Vectorize

vectorize-io/vectorize-mcp-server

✓SAFE B

Official Vectorize MCP Server

★ 1153 readstdio

Medical

jamesanz/medical-mcp

✓SAFE B

An MCP server that provides comprehensive medical information by querying multiple authoritative medical APIs including FDA, WHO, PubMed, Google Scholar, and RxNorm

★ 11417 readstdio

medical-mcp

JamesANZ/medical-mcp

✓SAFE B

An MCP server that provides comprehensive medical information by querying multiple authoritative medical APIs including FDA, WHO, PubMed, Google Scholar, and RxNorm

★ 11417 readstdio

Meting Agent

eldment/meting-agent

✓SAFE B

🪐 面向 AI 的多音乐平台的 API 代理 • 网易云 • QQ音乐 • 酷狗 • 酷我 🎻 Multi-platform music API agent for AI • NetEase • QQ Music • KuGou • Kuwo

★ 1121 readstdio

Korean DART

chrisryugj/korean-dart-mcp

✓SAFE B

OpenDART 전자공시MCP | 83개 API → 15 MCP 도구. 공시·재무·지분·XBRL + 버핏급 애널리스트 프레임(내부자 시그널·회계 리스크·퀄리티 체크리스트) + HWP/PDF 첨부 마크다운화 | 83 Korean DART disclosure APIs → 15 MCP tools with Buffett-grade analyst frames

★ 11018 readstdio

Screenshot

just-every/mcp-screenshot-website-fast

✓SAFE B

Quickly screenshots webpages and converts to an LLM friendly size

★ 1104 readstdio

Other risk hubs

By job

Questions

What does read-only mean here?
A server is listed here when its source declares at least one tool and every tool is classed read: no write tools and no destructive tools.
Is a read-only server always safe?
No. 913 are SAFE, 261 CAUTION and 446 BLOCK on this list. A tool list says what the tools do, not whether the surrounding code is clean, so a read-only server can still be BLOCK.
Why are some servers on no risk list?
615 servers declare no tools we can read, because they build the list at runtime. They cannot be called read-only, so they appear on none of the risk hubs.

Keep going in your AI

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