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/61

Verdicts · 3,151 servers

✓SAFE1,634!CAUTION522✕BLOCK995

Transports

A server can list more than one transport.

3,151 servers · 1,441–1,464 · ranked by trust score, then stars
create-ai-memory!CAUTION B

rambaarde/create-ai-memory

Persistent, agent-agnostic session memory for AI coding CLIs (Claude Code, Codex, Gemini, Cursor, opencode). One Markdown vault, any agent.

★ 3GitHub stars2 write
email-design-mcp!CAUTION B

hema9265/email-design-mcp

Generate responsive, professional email templates using AI with brand memory, expert design, live preview, and easy editing in MCP-compatible clients.

★ 3GitHub stars3 writestdio
josesepulvedapino-zahori!CAUTION B

josesepulvedapino/zahori

Find the media stream behind any web page: browser-driven HLS/DASH/MP4 resolution with learnable site profiles

★ 3GitHub stars1 writestdio
katto-mcp✓SAFE B

miracleweasel/katto-mcp

Katto's official MCP server for AI video clipping — turn long videos & podcasts into scored, captioned 9:16 clips from Claude, Cursor or ChatGPT.

★ 3GitHub stars1 writestdio
korean-dart-mcp✓SAFE B

aesthetic-legalism5470/korean-dart-mcp

Access OpenDART financial data and corporate filings via MCP. Analyze disclosures, financial statements, and XBRL data directly in your AI assistant.

★ 3GitHub stars15 readstdio
oorep-mcp✓SAFE B

Dhi13man/oorep-mcp

Model Context Protocol (MCP) server and client SDK that provides AI assistants with access to OOREP (Open Online Repertory) - a comprehensive homeopathic repertory and materia medica database.

★ 3GitHub stars1 writestdio
promptwheel-ai-logbook✓SAFE B

promptwheel-ai/logbook

Turn Git history into memory AI coding agents can use.

★ 3GitHub stars5 readstdio
release-intel-mcp✓SAFE B

barissozudogru/release-intel-mcp

Auditable release evidence from commits, pull requests, issues, and contributors between two Git refs, over MCP or CLI.

★ 3GitHub stars3 readstdio
stanislawherjan1-gdocs-comments✓SAFE B

stanislawherjan1/gdocs-comments-mcp

MCP server for adding inline, range-anchored comments to Google Docs — the one comment operation the Google Docs & Drive APIs can't do. For AI agents (Claude, Cursor, Copilot) that review documents.

★ 3GitHub stars1 writestdio
string-ai-mcp✓SAFE B

usestring/string-ai-mcp

Model Context Protocol server for the Web Access API.

★ 3GitHub stars5 readstdio
tanilo-mcp✓SAFE B

TKCollective/tanilo-mcp

MCP server for Tanilo (formerly AgentOracle): connects MCP clients to the Tanilo API. npm package: agentoracle-mcp.

★ 3GitHub stars3 readstdio
test-intel-mcp✓SAFE B

barissozudogru/test-intel-mcp

Local test coverage intelligence for JavaScript and TypeScript: gaps, complexity, and practical test priorities over MCP or CLI.

★ 3GitHub stars4 readstdio
Adguardctrl✓SAFE B

lidless-labs/adguardctrl

One bad call can take down your DNS filtering. adguardctrl gates every AdGuard Home write behind confirm and destructive flags. CLI plus MCP adapter, no daemon.

★ 2GitHub stars1 readstdio
comparedge-llm-cost✓SAFE B

comparedge/llm-cost-mcp

MCP server: token cost math for LLM API calls, 69 models across 17 providers, prices verified by ComparEdge

★ 2GitHub stars7 readstdio
comparedge-price-watch✓SAFE B

comparedge/price-watch-mcp

MCP server: verified SaaS price-change history, watch-log for AI assistants, data by ComparEdge

★ 2GitHub stars6 readstdio
dsh-verify✓SAFE B

263311487-ux/dsh-verify

Browser acceptance checks for agent-built web apps. Zero-key starter, CLI, MCP and GitHub Action; inspect explicit assertions and reports.

★ 2GitHub stars2 writestdio
facebook-ad-library-mcp-cli✓SAFE B

thenavidm/facebook-ad-library-mcp-cli

Facebook Ad Library MCP server and CLI for Claude Code, Codex and AI agents. 9 tools that read every ad running on Facebook, Instagram, Messenger, Threads and Audience Network, for any advertiser in any country, free and with no API key.

★ 2GitHub stars2 write
firmauy-mcp-inspect✓SAFE B

carlosplanchon/firmauy-mcp-inspect

Read-only MCP server that brings FirmaUY signature inspection to your AI assistant. Offline, no smart card, no signing, redacted by default.

★ 2GitHub stars4 read
gdharness✓SAFE B

Aureliolo/gdharness

A harness for driving a Godot 4 project from an agent: editor addons, a runtime bridge into the running game, an MCP server in front of them, and a CLI that installs and diagnoses the Godot side.

★ 2GitHub stars4 writestdio
hubmesh✓SAFE B

DemigodDSK/hubmesh

Centrality-aware GraphRAG retrieval planner — drop-in layer over any vector DB. Zero LLM in the query path; MCP server included.

★ 2GitHub stars8 readstdio
Layout-debug-mcp✓SAFE B

AntonChuraev99/Layout-debug-mcp

Point at a layer in a live UI (web DOM, Android Compose), drag it on the real screen, hand it to your AI agent via MCP

★ 2GitHub stars6 readstdio
mcp-atlassian✓SAFE B

Vijay-Duke/mcp-atlassian

MCP server for seamless Atlassian integration (Confluence & Jira) with AI assistants

★ 2GitHub stars8 writestdio
mcp-eval✓SAFE B

cavi-ai/mcp-eval

🧐 Privacy-preserving MCP server evaluation: deterministic CI probe battery (18 probe kinds, readiness score, SARIF) plus production friction capture with findings that verify and close

★ 2GitHub stars1 readstdio
mcp-google-analytics✓SAFE B

leonardosepulvedat/mcp-google-analytics

GA4 MCP that reads AND writes: reports, funnels, audits, and server-side events (Measurement Protocol). 26 tools.

★ 2GitHub stars14 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 (1,622 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.