MCP Trust Registry

Every server below was scanned with the same deterministic engine that powers the MCP Trust Checker CLI and free API — an auditable A–F Trust Score, the server's capability blast radius, and every finding with its evidence. No pay-to-list, no self-reported ratings: the score is computed from the package's actually published code and tool surface.

18,058 servers 22,930 packages scanned 23 categories engine mcptrustcheckersame input → same score what changed →
Playwright (official) Official 18 implementations
A
best: @playwright/mcp

Microsoft’s official Playwright MCP server: drive a real browser through accessibility-tree snapshots.

Browser Automation Best score 100/100 0 findings 6.2M/wk Cap. Minimal
Transport Http Source verified npm
A
@tmcp/transport-http

Security scan results for the Transport Http MCP server.

API Development Score 100/100 0 findings 1.9M/wk Cap. Minimal
Session Manager Source verified npm
A
@tmcp/session-manager

Security scan results for the Session Manager MCP server.

AI, Memory & Reasoning Score 100/100 0 findings 1.9M/wk Cap. Minimal
Storybook Source verified 6 implementations
A
best: @storybook/mcp

Security scan results for the Storybook MCP server.

Data Science & ML Best score 100/100 0 findings 1.9M/wk Cap. Minimal
Tmcp Source verified npm
A
tmcp

Security scan results for the Tmcp MCP server.

Data Science & ML Score 100/100 0 findings 1.8M/wk Cap. Minimal
Orval npm
A
@orval/mcp

Security scan results for the Orval MCP server.

API Development Score 100/100 0 findings 1.6M/wk Cap. Minimal
Addon Source verified npm
A
@storybook/addon-mcp

Security scan results for the Addon MCP server.

Data Science & ML Score 100/100 1 finding 1.5M/wk Cap. High
Context7 Official npm
A
@upstash/context7-mcp

Up-to-date, version-specific library documentation injected straight into prompts.

Developer Tools Score 100/100 0 findings 1.1M/wk Cap. Minimal
Handler Official npm
A
mcp-handler

Security scan results for the Handler MCP server.

Data Science & ML Score 100/100 2 findings 626k/wk Cap. High
Fastmcp Source verified 2 implementations
A
best: fastmcp

Security scan results for the Fastmcp MCP server.

Design & Media Best score 100/100 0 findings 535k/wk Cap. Minimal
Remote 4 implementations
A
best: mcp-remote

Security scan results for the Remote MCP server.

API Development Best score 100/100 0 findings 474k/wk Cap. Minimal
Filesystem (official) Official 24 implementations
A
best: @modelcontextprotocol/server-filesystem

The official MCP filesystem server: read, write, search and manage local files with configurable access roots.

Files & Storage Best score 100/100 0 findings 469k/wk Cap. Minimal
Mcporter 2 implementations
A
best: mcporter

Security scan results for the Mcporter MCP server.

API Development Best score 100/100 10 findings 443k/wk Cap. High
Hono Source verified 3 implementations
A
best: @hono/mcp

Security scan results for the Hono MCP server.

API Development Best score 100/100 0 findings 308k/wk Cap. Minimal
Adapters Official 2 implementations
A
best: @langchain/mcp-adapters

Security scan results for the Adapters MCP server.

AI & Agents Best score 100/100 4 findings 228k/wk Cap. High
Nest Source verified npm
A
@rekog/mcp-nest

Security scan results for the Nest MCP server.

Content Management Score 100/100 0 findings 216k/wk Cap. Minimal
Supabase (official) Official 12 implementations
A
best: @supabase/mcp-server-supabase

Supabase’s official MCP server: manage projects, run SQL and inspect your database.

Databases Best score 100/100 0 findings 185k/wk Cap. Minimal
Sequential Thinking (official) Official 10 implementations
A
best: @modelcontextprotocol/server-sequential-thinking

The official structured-reasoning server: break problems into revisable thought sequences.

AI, Memory & Reasoning Best score 100/100 0 findings 150k/wk Cap. Minimal
Tools Source verified npm
A
@clerk/mcp-tools

Security scan results for the Tools MCP server.

AI & Agents Score 100/100 0 findings 118k/wk Cap. Minimal
Gitlab Source verified 24 implementations
A
best: @zereight/mcp-gitlab

Security scan results for the Gitlab MCP server.

AI & Agents Best score 100/100 16 findings 115k/wk Cap. High
MongoDB (official) Official 10 implementations
A
best: mongodb-mcp-server

MongoDB’s official MCP server: query collections, inspect schemas and run aggregations.

Databases Best score 100/100 4 findings 101k/wk Cap. High
Memory (official) Official 34 implementations
A
best: @modelcontextprotocol/server-memory

The official knowledge-graph memory server: persistent entities, relations and observations.

AI, Memory & Reasoning Best score 100/100 0 findings 94k/wk Cap. Minimal
Figma Developer Source verified 2 implementations
A
best: figma-developer-mcp

Security scan results for the Figma Developer MCP server.

AI & Agents Best score 100/100 0 findings 86k/wk Cap. Minimal
Nx Source verified 2 implementations
A
best: nx-mcp

Security scan results for the Nx MCP server.

AI & Agents Best score 100/100 0 findings 81k/wk Cap. Minimal

A security-scanned MCP server directory, not just another list

The MCP Trust Registry catalogs Model Context Protocol servers — the connectors that let AI assistants such as Claude, Cursor and other LLM clients reach real systems: databases, browsers, cloud and DevOps tooling, memory stores and web search. Most MCP directories only aggregate links and download counts. Here, every listed server has first been analyzed by the open-source mcptrustchecker engine: the published npm or PyPI package is scanned deterministically, the same input always yields the same result, and nobody can pay to be listed or ranked higher.

How to pick a trusted MCP server using three axes

Each entry exposes three signals worth reading together before you install anything:

  • Trust Score — an A-F grade from 0-100 summarizing findings in the code and metadata; treat it as a first filter, then open the individual findings and their evidence.
  • Capability profile — the "blast radius": whether the server can touch the filesystem, spawn processes, or make network calls, and how those combine.
  • Coverage — how much of the package the scanner could actually analyze, which tells you how much weight the grade deserves.

Prefer the narrowest capability set that still does the job: an assistant that only needs read-only analytics has no reason to run a server with shell access. Anything not yet listed can be checked on demand through the free scan API.

Why security matters for Model Context Protocol servers

An MCP server sits inside the trust boundary of your AI assistant, so its risks differ from ordinary dependencies. Tool poisoning hides instructions in tool descriptions that the model reads and obeys. Prompt injection arrives through fetched pages, tickets or emails and can redirect the assistant mid-task. And an over-permissioned server turns either attack into data exfiltration: a connector that can both read your files and make outbound requests can quietly ship secrets to an attacker. Scores here are automated opinions derived from published package code — evidence to weigh, not certifications.

Is a high Trust Score a guarantee that an MCP server is safe?

No. A strong grade means the deterministic scan found no dangerous patterns in the published package at the scanned version, with the stated coverage. New releases change code, so re-check after updates, review the capability profile against your own threat model, and grant each server the least access it needs.