Stop Searching. Start Trusting.

The curated directory of MCP servers, vetted for security, efficiency, and quality.

Browse DirectoryWhy Us?

Tired of the MCP "Marketplace" Chaos?

We built MCPScout.ai to solve the ecosystems biggest pain points.

No Insecure Dumps

We manually analyze every server for basic security flaws.

Easy Setup

Our gotcha notes warn you about complex setups.

Avoid "Token Hogs"

We estimate token costs for cost-effective agents.

Products, Not Demos

We filter out "Hello World" demos.

CATEGORIES:
SORT:

Vetted Servers(9120)

0
0
Low Cost
paularlott icon

mcp

by paularlott

Sec8

A Go library for building Model Context Protocol (MCP) servers with a fluent API, especially for integrating with Large Language Models (LLMs) to provide and execute tools efficiently.

Setup Requirements

  • ⚠️LLM integration (via `openai` subpackage) typically requires an OpenAI API Key (or equivalent for other LLMs), which can incur costs.
  • ⚠️The `openai` subpackage's examples are configured to use LM Studio (a local LLM server) running on `127.0.0.1:1234` with a specific model (`qwen/qwen3-1.7b`). This local setup is required for the examples to work out-of-the-box.
Verified SafeView Analysis
The library adheres to good Go practices, utilizing standard networking and JSON handling with internal error handling for invalid requests/parameters. Key security features include default `InsecureSkipVerify: false` for TLS, and `MAX_TOOL_CALL_ITERATIONS` to prevent infinite tool loops in LLM integration. Potential risks typically stem from user-implemented tool handlers (e.g., unsafe file I/O or OS command execution without validation) rather than the core library itself. While tool search could be a DoS vector for very long inputs if not protected by an upstream API gateway, the internal implementation does not show immediate vulnerabilities.
Updated: 2026-01-17GitHub→
0
0
Medium Cost

ynab-mcp

by doctor-kat

Sec6

A Node.js and TypeScript project designed to extend or interact with the YNAB (You Need A Budget) budgeting application.

Setup Requirements

  • ⚠️Node.js runtime environment required.
  • ⚠️pnpm package manager required for dependency management.
  • ⚠️Likely requires a YNAB API token for full functionality.
  • ⚠️Potentially requires an Anthropic (Claude) API key if AI is used at runtime.
Review RequiredView Analysis
No source code was provided for a thorough security analysis; assessment is based on file names only. Risks related to 'eval', code obfuscation, or specific network vulnerabilities cannot be determined. Potential risks include improper handling of API keys or lack of input validation typical for web services.
Updated: 2025-11-18GitHub→
0
0
Medium Cost
CR-AudioViz-AI icon

crav-mcp-vercel

by CR-AudioViz-AI

Sec8

Vercel deployment automation and resource management for AI applications, enabling autonomous deployments, real-time build monitoring, log retrieval, error parsing, and resource management.

Setup Requirements

  • ⚠️Requires a Vercel access token (`VERCEL_TOKEN`) with deployment permissions.
  • ⚠️Requires a secure, user-defined API key (`MCP_API_KEY`) for this server's authentication.
Verified SafeView Analysis
The server employs standard security middleware (`helmet`, `cors`, `express-rate-limit`) and uses API key authentication (`MCP_API_KEY`) for all sensitive endpoints. The Vercel access token (`VERCEL_TOKEN`) is correctly sourced from environment variables. No obvious hardcoded secrets or malicious patterns were found in the core server logic. The primary security relies on the confidentiality and integrity of the `MCP_API_KEY` and `VERCEL_TOKEN`.
Updated: 2026-01-09GitHub→
0
0
High Cost
peakmojo icon
Sec7

Automate software development tasks using Claude Code AI agents within isolated E2B sandboxes, providing a fixed set of skills and a job-level API.

Setup Requirements

  • ⚠️Requires E2B API Key (Paid service)
  • ⚠️Requires Claude API Access (Paid service, implied by 'Claude Code')
  • ⚠️Assumes a pre-built E2B base image with Claude Code installed and configured
Review RequiredView Analysis
The system design leverages E2B for sandbox isolation, which is a strong security principle for running arbitrary code. Skills are bundled and fixed, reducing injection risks. However, the actual server-side code (MCP implementation) for provisioning sandboxes, handling client input, managing secrets (like E2B/Claude API keys), and securely copying/mounting user-provided repositories is not available for audit. Without reviewing the implementation, critical vulnerabilities related to input validation, secret management, or sandbox escape cannot be ruled out. The 'eval' keyword, obfuscation, or hardcoded secrets cannot be checked without the full source code. The security score reflects the robust design intent, not an audited implementation.
Updated: 2025-12-02GitHub→
0
0
Medium Cost
eagleisbatman icon

traduora-mcp-server

by eagleisbatman

Sec8

An MCP server for managing translations in Traduora via its API, acting as a tool for a larger model or system.

Setup Requirements

  • ⚠️Requires a running Traduora instance (e.g., deployed on Railway with a MySQL database).
  • ⚠️Requires generation and configuration of OAuth2 client ID and client secret within Traduora for the MCP server to authenticate.
  • ⚠️Critical environment variables (TRADUORA_API_URL, TRADUORA_CLIENT_ID, TRADUORA_CLIENT_SECRET) must be set for full functionality; a warning is logged if missing.
Verified SafeView Analysis
The server correctly uses environment variables for sensitive configurations (Traduora API URL, client ID, client secret). It employs `zod` for robust input validation of tool arguments, mitigating common injection risks. CORS is enabled, defaulting to '*' if `ALLOWED_ORIGINS` is not explicitly set in environment variables, which is a potential security risk in a production environment if not configured to restrict origins. There are no direct uses of `eval` or obvious malicious patterns in the provided source code. The OAuth2 client credentials flow is standard.
Updated: 2025-11-29GitHub→
0
0
Medium Cost
manuel2f icon

mcp-servers-manu

by manuel2f

Sec8

This server provides tools for searching and exploring internal CDN documentation, including specific components, DTS, and Wiki content.

Setup Requirements

  • ⚠️Requires access to an internal CDN documentation server at `http://cdn-docs.cdn.hi.inet` (or a custom URL via `CDN_DOCS_BASE_URL`).
  • ⚠️Requires Node.js and npm installed to run the TypeScript server.
  • ⚠️For full integration, it needs to be configured in a Model Context Protocol (MCP) client like VS Code's `mcp.json`.
Verified SafeView Analysis
The server primarily functions as a web scraper for internal documentation, using `axios` and `cheerio`. The default `BASE_DOCS_URL` points to an internal network address (`cdn-docs.cdn.hi.inet`), limiting external exposure. No `eval` or dynamic code execution is observed, and no hardcoded secrets are present. The main security considerations would be ensuring the internal documentation source itself is trusted and not susceptible to content injection that could bypass `cheerio`'s sanitization, or potential SSRF if `CDN_DOCS_BASE_URL` is set to an untrusted external URL via environment variable.
Updated: 2025-12-01GitHub→
0
0
Medium Cost
jorgearturoyap-debug icon

metatrader-5-for-Chatgpt-Desktop

by jorgearturoyap-debug

Sec7

A Model Context Protocol (MCP) server for MetaTrader 5, specifically adapted for ChatGPT Desktop integration to enable trading and market data analysis.

Setup Requirements

  • ⚠️Requires a ChatGPT Plus subscription or higher (Free tier does not support MCP).
  • ⚠️Requires MetaTrader 5 terminal installed on Windows, limiting its use to Windows OS.
  • ⚠️Requires ngrok setup (account, auth token) for public tunneling, and the dynamically generated ngrok HTTPS URL must be manually updated in ChatGPT Desktop settings upon each new ngrok session (free ngrok sessions expire frequently).
Verified SafeView Analysis
The server explicitly warns that it is publicly exposed via ngrok and recommends setting an `MCP_API_KEY` for authentication. If `MCP_API_KEY` is not configured, the server is open to all connections, which is a significant security risk. The `start_chatgpt.bat` script binds the server to `0.0.0.0` (all interfaces) which is necessary for ngrok but requires careful network security. There are no 'eval' or obvious malicious patterns, and the code relies on the official MetaTrader5 Python API. Users are strongly advised to use a demo account first and configure the API key.
Updated: 2025-11-19GitHub→
0
0
Medium Cost
Aliimran14 icon

prop_crm

by Aliimran14

Sec1

A server component for property customer relationship management, designed to integrate with a GitHub MCP (Microservice Communication Protocol) environment.

Review RequiredView Analysis
Crucial source code (application logic, dependencies, server configuration, etc.) was not provided beyond the README.md. Therefore, a comprehensive security audit for 'eval', obfuscation, network risks, hardcoded secrets, or malicious patterns is impossible. The low score reflects an inability to verify safety due to missing code, rather than identified vulnerabilities.
Updated: 2025-12-05GitHub→
0
0
Low Cost
darkdragonsastro icon

dda-mcp-server

by darkdragonsastro

Sec8

The DDA MCP Server controls astronomy equipment, including proprietary Dark Dragons Astronomy devices and any ASCOM Alpaca-compatible hardware, by exposing their functionalities as Model Context Protocol (MCP) tools for agent interaction.

Setup Requirements

  • ⚠️Requires Node.js and npm/npx to run.
  • ⚠️Requires physical Dark Dragons Astronomy devices or ASCOM Alpaca-compatible hardware on the local network to be functional.
  • ⚠️ASCOM Alpaca devices must be explicitly connected using the `alpaca_device_connect` tool before other operations can be performed.
Verified SafeView Analysis
The server uses `fetch` to interact with user-provided IP addresses and ports on the local network for device control and discovery. While inherent to its function of controlling local astronomy equipment, this pattern could pose a Server-Side Request Forgery (SSRF) risk if the MCP server itself were exposed to untrusted external requests. However, given its intended execution context (local MCP server via stdio, often by an agent like Claude Desktop), the risk is mitigated as it primarily interacts with trusted local network devices. No `eval`, code obfuscation, or hardcoded secrets were found.
Updated: 2025-11-27GitHub→
0
0
Medium Cost
WilliamPinto-Olmos icon

project-mcp-server

by WilliamPinto-Olmos

Sec7

Provides a Model Context Protocol server to allow LLM agents to explore and interact with a project's APIs (via OpenAPI) and databases (currently MySQL).

Setup Requirements

  • ⚠️Requires a local OpenAPI specification file (defaults to `openapi-spec.json` in CWD, or specified by `PROJECT_MCP_OPENAPI_SPEC`).
  • ⚠️Database features require MySQL/MariaDB database credentials (`PROJECT_MCP_DB_HOST`, `PROJECT_MCP_DB_USER`, `PROJECT_MCP_DB_PASSWORD`, `PROJECT_MCP_DB_DATABASE`).
  • ⚠️Intended for use with a Model Context Protocol (MCP) compatible client (e.g., Cursor, Claude Desktop), requiring specific client-side configuration.
Verified SafeView Analysis
The server dynamically generates tools for API and database interaction. For APIs, it uses axios with an authentication context, which is standard. For databases, it employs a `SqlValidator` to restrict query types (e.g., read-only, update-only, delete-only) and block dangerous SQL keywords before execution. This is a good layer of defense. Crucially, the most dangerous database tool, `db_run_statement` (allowing arbitrary SQL), is explicitly disabled by default and requires an opt-in via environment variable or programmatic configuration, with clear warnings. However, relying on keyword blocking for SQL validation, while implemented with care (stripping comments/strings and tokenizing), is not 100% foolproof against highly sophisticated or novel SQL injection techniques by a sufficiently capable or adversarial LLM. Users must be cautious with the `enableRunDeleteStatement` and especially `enableRunStatement` permissions.
Updated: 2026-01-19GitHub→
0
0
Low Cost
GlacierEQ icon
Sec9

This MCP server provides specialized legal tools to analyze federal civil rights cases, detect judicial bias, track evidence, monitor court deadlines, and assess child welfare for Case 1FDV-23-0001009.

Setup Requirements

  • ⚠️Requires Node.js v18.0.0 or higher to run.
Verified SafeView Analysis
The server appears secure. It uses standard I/O for communication via the Model Context Protocol SDK, reducing typical network attack surfaces. There are no instances of `eval` or dynamic code execution found. No hardcoded sensitive secrets or API keys are present in the provided source code. Input handling for tool arguments is structured and responses are generated from predefined templates, minimizing injection risks. The `caseId` is hardcoded as a case identifier, not a credential.
Updated: 2025-11-23GitHub→
0
0
High Cost
Pranavashokk icon
Sec2

Manages and automates image generation requests, incorporating a cost approval workflow for bulk operations to manage long-running tasks.

Setup Requirements

  • ⚠️Requires an operational 'MCP server' for the agent to connect to.
  • ⚠️Requires API access to an image generation service (e.g., OpenAI DALL-E, Stability AI), which will incur external costs.
  • ⚠️Likely requires an API key for the image generation service, which needs to be configured securely.
Review RequiredView Analysis
The provided content is limited to a README file; actual source code for the server and agent was not supplied for analysis. Therefore, a comprehensive security audit for 'eval', obfuscation, network risks, hardcoded secrets, or malicious patterns is impossible. The low score reflects the inability to verify security without the actual code.
Updated: 2025-12-14GitHub→
PreviousPage 690 of 760Next