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
spre-sre icon

mcpm

by spre-sre

Sec5

A CLI tool for developers to install, manage, and register Model Context Protocol (MCP) servers with Claude Code and Gemini CLI.

Setup Requirements

  • ⚠️Requires Go 1.23+ to build from source.
  • ⚠️Requires Git to clone repositories for installation.
  • ⚠️Requires Node.js/npm, Python 3, and/or Go runtime environments on the host machine if installing servers of those types.
  • ⚠️Requires Claude Code CLI and/or Gemini CLI to register installed MCP servers with those clients.
Review RequiredView Analysis
The tool's core functionality involves downloading and executing arbitrary code (MCP servers) from user-specified or inferred repositories. This is an inherent risk for any package manager. Specifically, the `internal/builder/shell.go` executes build commands using the user's shell (`zsh -l -c command` or `bash -l -c command`), which can be more susceptible to shell injection if the command strings themselves were derived from untrusted, unsanitized user input. The `mcpm add` command allows direct registration of arbitrary commands and arguments for MCP servers. While this is an intended feature, it means the security responsibility lies entirely with the user to ensure the commands and repositories they instruct `mcpm` to handle are trusted. No direct hardcoded secrets or obfuscation were identified within the provided source code, but the execution of untrusted external code remains a significant vector for security compromise.
Updated: 2025-12-10GitHub→
0
0
Medium Cost
franco-gh icon

az-mcp-tf

by franco-gh

Sec9

Deploys the terraform-mcp-server to Azure Container Apps, providing an interface for Terraform interaction compatible with clients like GitHub Copilot Chat.

Setup Requirements

  • ⚠️Requires Terraform and Azure CLI installation and configuration.
  • ⚠️Requires Azure subscription with Contributor access and ability to create RBAC role assignments.
  • ⚠️Requires specific GitHub Secrets and Variables for automated CI/CD deployment via GitHub Actions and Terraform Cloud.
  • ⚠️Potential Key Vault naming conflicts requiring manual purging of soft-deleted vaults.
Verified SafeView Analysis
The project is designed with strong security practices, including API key authentication, Azure Key Vault for secret storage, Managed Identities for secure access, RBAC for Key Vault access, and HTTPS by default for all communication. No 'eval' or obvious obfuscation detected. API keys are not hardcoded. The architecture places Nginx in front for API key validation, enhancing security.
Updated: 2025-12-02GitHub→
0
0
Medium Cost
AnotherStream icon

mcp-notes-connector

by AnotherStream

Sec9

Implements a Model Context Protocol (MCP) server to integrate with the Evernote API, allowing MCP clients (like AI agents) to access and manage Evernote notes and notebooks.

Setup Requirements

  • ⚠️Requires an Evernote Developer Token (needs registration on dev.evernote.com)
  • ⚠️Requires Python 3.10 or higher
  • ⚠️Testing is strongly recommended in the Evernote Sandbox environment to avoid affecting real data
Verified SafeView Analysis
The project uses environment variables for sensitive data (EVERNOTE_TOKEN) via `python-dotenv`, which is good practice. MCP communication occurs over `stdio`, limiting direct network exposure locally. The `evernote3` SDK is used for external API calls, which is generally reliable. No `eval` or `exec` found. Full input validation for currently unimplemented API methods (`search_notes`, `get_note`, `create_note`) needs to be confirmed once they are implemented, but the current structure handles arguments safely.
Updated: 2025-11-23GitHub→
0
0
Medium Cost
BharatDeepanTR icon
Sec8

An AI-powered IT Helpdesk for Thomson Reuters employees, integrating real TR service desk resources and Claude AI for enhanced support.

Setup Requirements

  • ⚠️Requires access to a specific AWS account (818565325759) with the pre-existing IAM role 'a208194-askjulius-agentcore-gateway' and Lambda function 'a208194-it-helpdesk-enhanced-mcp-server' already configured.
  • ⚠️Deployment script (`deploy-enhanced-mcp-cloudshell.sh`) is optimized for AWS CloudShell, although it might run in other environments.
  • ⚠️The Lambda runtime is explicitly set to Python 3.9 in the deployment script, conflicting with the Python 3.14 mentioned in the README. Python 3.9 compatibility is required for the Lambda function code.
Verified SafeView Analysis
The code uses standard AWS SDK calls for Lambda invocation and Bedrock interactions, which rely on configured AWS credentials for authentication. There are no direct hardcoded API keys or sensitive secrets exposed in the provided code snippets. The deployment script performs URL validation using `curl`, which is a common and generally safe practice. Hardcoded AWS resource identifiers (Lambda name, IAM role ARN, Agent ID) are present but are configuration details rather than security secrets. The `bedrock_agentcore.memory` SDK is conditionally imported, with a fallback if not available, mitigating direct dependency risks. No 'eval' or other highly dangerous patterns were observed. The overall security posture is good for an internal enterprise tool, though externalizing all configuration could improve flexibility.
Updated: 2025-12-04GitHub→
0
0
High Cost
liquid1982 icon
Sec9

Provides read-only access to the TrustLayer API via the Model Context Protocol (MCP) for integration with AI clients and other MCP-compatible tools.

Setup Requirements

  • ⚠️Requires Python 3.11+
  • ⚠️Requires a valid TrustLayer API token for authentication
Verified SafeView Analysis
The server is explicitly designed for read-only access (GET operations only), significantly reducing the attack surface. It properly handles API tokens via environment variables or per-request Authorization headers using Python's contextvars, preventing hardcoded secrets. There is no evidence of `eval` or code obfuscation. Uses well-known libraries (`httpx`, `pydantic-settings`). Production deployment notes emphasize HTTPS and firewalling, which are critical for any exposed service. The only minor deductions would be for standard deployment considerations (e.g., ensuring secure infrastructure).
Updated: 2025-12-05GitHub→
0
0
Medium Cost
TheaMarieM icon
Sec8

AI-powered portfolio acting as a digital twin, offering interactive querying, interview simulation, and RAG-driven semantic search of professional experiences via the Model Context Protocol (MCP) for AI assistants like Claude Desktop and VS Code Copilot.

Setup Requirements

  • ⚠️Requires Node.js 18+ and pnpm
  • ⚠️Requires Python 3.8+ with pip
  • ⚠️Requires Upstash accounts (Redis + Vector) and corresponding API keys for full functionality
  • ⚠️Requires a Groq API key for LLM responses (or optional OpenAI API Key fallback/embeddings, depending on configuration)
  • ⚠️Requires a local Ollama server running if `LLM_PROVIDER` is 'ollama'
  • ⚠️Requires a local embedding service running (`scripts/serve_local_embeddings.py`) if `USE_LOCAL_EMBEDDINGS` is 'true'
Verified SafeView Analysis
The project demonstrates good security practices with explicit rate limiting (Redis-based for chat, in-memory for RAG as implemented in `app/api/rag/route.ts`), comprehensive input validation, and content filtering for sensitive information and prompt injections. It uses secure session ID generation and sets security headers on API responses. Environment variables are correctly used for secrets. The use of `json.loads` on external API responses carries inherent risk, but is standard practice when interacting with LLM/embedding services and is handled within `try-except` blocks. In-memory rate limiting for RAG in the Next.js backend could be a concern for multi-instance production deployments, which the README addresses by suggesting Redis for production.
Updated: 2025-11-19GitHub→
0
0
Low Cost
bioanywhere icon

tacos

by bioanywhere

Sec7

Provides Docker deployment configuration and CI/CD for a world-news-api-clients server.

Setup Requirements

  • ⚠️Requires Docker for local development and execution.
  • ⚠️Requires a GCP account, project ID, and project number for cloud deployment and Artifact Registry integration.
  • ⚠️Requires GitHub repository secrets (`GCP_PROJECT_ID`, `GCP_PROJECT_NUMBER`) for CI/CD.
  • ⚠️Requires `gcloud` CLI installed and configured to pull images from Google Artifact Registry.
Verified SafeView Analysis
The provided source code (README) outlines secure practices for GCP integration, utilizing Workload Identity Federation and repository secrets for credentials. There are no obvious hardcoded secrets or malicious patterns visible in the README itself. However, a comprehensive security audit for `eval`, obfuscation, network risks within the application, or insecure configurations is not possible without access to the actual `Dockerfile`, `docker-compose.yml`, `run.sh`, and the application's source code.
Updated: 2025-12-13GitHub→
0
0
High Cost
adamanz icon
Sec8

An MCP server enabling AI agents like Claude to analyze videos and images using the Qwen3-VL-8B-Instruct model deployed on Blaxel's H100 GPUs.

Setup Requirements

  • ⚠️Requires a Blaxel account and API key (Blaxel is a paid, subscription-based service).
  • ⚠️Requires `ffmpeg` to be installed locally on the system where the MCP server runs.
  • ⚠️Requires Python 3.10 or newer.
  • ⚠️Requires manual configuration in Claude Desktop's `claude_desktop_config.json`.
Verified SafeView Analysis
The server uses `subprocess.run` to execute `ffmpeg` for video frame extraction. While `ffmpeg` is a standard and generally robust tool, processing external video URLs with it can theoretically expose the system to vulnerabilities if a maliciously crafted video file exploits a flaw in `ffmpeg`. The code itself does not show immediate signs of code injection or hardcoded secrets, and API keys are loaded from environment variables. Network requests are handled by `httpx` with timeouts.
Updated: 2025-11-30GitHub→
0
0
High Cost
firda0802 icon
Sec9

An AI-powered platform for generating and managing various types of documents, presentations, spreadsheets, creative content (ebooks, stories), and multimedia (images, videos, voiceovers), integrated with chat capabilities.

Setup Requirements

  • ⚠️Requires a Google Gemini API Key (via Lovable AI Gateway)
  • ⚠️Requires an ElevenLabs API Key for voice generation
  • ⚠️Requires a Firebase Project setup (API Key, Auth Domain, Project ID, etc.)
  • ⚠️Requires a Supabase Project setup (URL, Publishable Key, Service Role Key, Database Schema, RPC functions)
  • ⚠️Requires a Stripe account and API keys for subscription management and payments
  • ⚠️Requires a Resend API Key for transactional emails
  • ⚠️Supabase Edge Functions run on Deno, requiring Deno for local development/testing of functions.
Verified SafeView Analysis
The project demonstrates strong security practices by utilizing environment variables for all sensitive API keys and secrets, robust Firebase authentication for user management, and Supabase Edge Functions as an authenticated API gateway. Client-side input is sanitized (e.g., HTML content via DOMPurify). Server-side rate limiting and usage checks are implemented to prevent abuse. No obvious malicious patterns or unsafe `eval` usage were detected. The primary area for external risk is the reliance on third-party APIs (Gemini/Lovable, ElevenLabs, Stripe, Resend), whose security postures are beyond the direct control of this application.
Updated: 2026-01-19GitHub→
0
0
Medium Cost

DemoProject

by sathyamoorthysps-tech

Sec5

Develop and execute automated web browser tests using the Behavior-Driven Development (BDD) methodology with Playwright.

Setup Requirements

  • ⚠️Node.js environment required.
  • ⚠️Requires installation of Playwright browser dependencies (often automated, but can be manual).
  • ⚠️An application or website to test is necessary for practical use of the framework.
Review RequiredView Analysis
No code or repository content was provided to audit for 'eval', obfuscation, or specific network risks. The project name suggests a web testing framework, which typically involves browser interactions and network requests to target applications. Without code, safety cannot be guaranteed.
Updated: 2025-11-18GitHub→
0
0
Medium Cost
Digital-Defiance icon

mcp-recording

by Digital-Defiance

Sec1

Provides recording and playback functionality for an MCP server, likely integrated within a larger AI capabilities suite.

Review RequiredView Analysis
Security cannot be assessed as only the README.md file was provided as 'SOURCE CODE'. Without access to the actual server implementation files, it's impossible to check for critical vulnerabilities such as 'eval' usage, obfuscation, network risks, hardcoded secrets, or malicious patterns. Therefore, it is considered highly unsafe to run without a full code audit.
Updated: 2025-12-04GitHub→
0
0
Medium Cost
Rodr116 icon
Sec9

This MCP server provides AI assistants with direct access to the Gentleman Programming Book for reading, searching, and understanding software architecture concepts.

Setup Requirements

  • ⚠️Requires the 'Gentleman Programming Book' repository to be cloned locally, and the 'BOOK_PATH' environment variable correctly set to its 'src/data/book' directory.
  • ⚠️Semantic search requires either a paid OpenAI API key (set as 'OPENAI_API_KEY') or a locally running Ollama server with the 'nomic-embed-text' embedding model (free, local).
  • ⚠️Requires Go 1.21 or superior to compile from source.
Verified SafeView Analysis
The server handles file system access by reading MDX book files from a configurable 'BOOK_PATH'. This path is set via an environment variable, which requires the user to ensure it points to a trusted location. API keys (OpenAI) and Ollama URLs are also configured via environment variables, which is good practice. There are no apparent 'eval' statements, command injection vulnerabilities, or malicious patterns found in the provided source code. External network calls are made to OpenAI or Ollama for embeddings, which are standard for such functionality.
Updated: 2026-01-19GitHub→
PreviousPage 295 of 760Next