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
Medium Cost
bajpainaman icon

DeltaMCP

by bajpainaman

Sec9

Enhances AI assistants with syntax-highlighted and colorized Git diffs, blame, and grep output via Model Context Protocol.

Setup Requirements

  • ⚠️Requires git-delta (installed via Cargo).
  • ⚠️Requires Git.
  • ⚠️Requires Python 3.10+.
  • ⚠️MCP client configuration requires absolute paths to the server directory.
  • ⚠️The client application must be fully restarted after configuration changes for the server to be recognized.
Verified SafeView Analysis
The server demonstrates strong security practices. Input validation is rigorously applied to file paths (preventing path traversal) and commit ranges (using regex patterns). Grep tools are whitelisted to prevent arbitrary command execution. Subprocess execution (`asyncio.create_subprocess_exec`) is used with command lists, not shell strings, mitigating command injection. The server operates over STDIO, minimizing network attack surface, and generated HTML files for browser viewing are local (`file://` URLs). No hardcoded secrets were found. Logging is directed to stderr to prevent interference with STDIO communication.
Updated: 2025-12-14GitHub→
0
0
Medium Cost
amberofficil icon

mcp_server_yt

by amberofficil

Sec1

Unable to determine the specific use case as no source code was provided for analysis.

Review RequiredView Analysis
CRITICAL: No source code was provided for analysis. Therefore, a thorough security audit for 'eval', obfuscation, network risks, hardcoded secrets, or malicious patterns could not be performed. Running this server without reviewing its code is extremely risky and is strongly advised against, as it may contain malicious patterns, hardcoded secrets, or unpatched vulnerabilities. The low score reflects this complete lack of visibility.
Updated: 2025-11-27GitHub→
0
0
Medium Cost
akbfar icon
Sec10

This application allows users to connect to a Zomato MCP Server via Claude Desktop to search for restaurants, browse menus, and place food orders using AI technologies.

Setup Requirements

  • ⚠️Requires downloading and installing an executable (.exe or .zip) from an external release page, which carries security risks without source code verification.
  • ⚠️Requires an internet connection for fetching restaurant data.
  • ⚠️Requires server details (API key and endpoint URL) from Zomato, to be entered in the application's settings.
Review RequiredView Analysis
The only 'source code' provided for analysis is the `Readme.md` file. This file itself contains no direct security risks, malicious patterns, `eval` statements, obfuscation, or hardcoded secrets. However, the `Readme.md` describes a desktop client application that requires users to download and install an executable (`.exe` or `.zip`) from a GitHub release page. A comprehensive security audit of the actual client application's binary logic or its source code (which was not provided) would be necessary to assess its real-world security posture. Running arbitrary executables without source code review carries inherent risks.
Updated: 2026-01-19GitHub→
0
0
Low Cost
mayorGonzalez icon

mcp-server-prueba

by mayorGonzalez

Sec9

Exposes VoltAgent AI agents and their integrated tools over the Model Context Protocol (MCP) using a Hono HTTP server, enabling discovery and invocation by compatible IDEs.

Setup Requirements

  • ⚠️Requires Node.js 20+ and pnpm.
  • ⚠️OPENAI_API_KEY must be configured in your environment for AI agent functionality.
  • ⚠️BUILDERBOT_API_KEY must be configured in your environment to interact with BuilderBot project management tools.
Verified SafeView Analysis
The server leverages established frameworks like VoltAgent, Hono, and Pino for logging, indicating good development practices. API keys (OPENAI_API_KEY, BUILDERBOT_API_KEY) are explicitly managed as environment variables, preventing hardcoding. Input validation is performed using Zod schemas for tools. Network exposure is primarily local (port 3141 by default) and via a well-defined protocol (MCP). The use of BuilderBotApiClient involves external API calls, which are handled with error checking. No obvious 'eval' or malicious code patterns were found in the provided snippets. The security score reflects confidence in the framework's design and a clear separation of concerns, while acknowledging that any networked application carries inherent risks.
Updated: 2025-11-30GitHub→
0
0
Medium Cost
pinchasn81 icon
Sec9

MCP server for LifeTracker DynamoDB integration, providing tools for activity logging and memory management related to health and daily life.

Setup Requirements

  • ⚠️Requires AWS Account and IAM User with DynamoDB permissions.
  • ⚠️Requires a Global Secondary Index (GSI) named 'user_name-index' on the 'ActivityLog' and 'MemoryEntry' tables for optimal performance; falls back to slower scan operations if not present.
  • ⚠️Known deployment issue with FastMCP cloud platform ('Already running asyncio in this thread' error); alternative deployment platforms like AWS Lambda or Docker are recommended.
  • ⚠️DynamoDB table names need to be configured via `TABLE_PREFIX` environment variable or directly match expected names ('ActivityLog', 'MemoryEntry').
Verified SafeView Analysis
Relies on environment variables for AWS credentials (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY), which must be securely managed in deployment environments. No 'eval' or other directly exploitable code patterns found. Data access is via standard DynamoDB operations.
Updated: 2025-11-27GitHub→
0
0
Low Cost
ansariwaqqas icon

storybook-mcp-server

by ansariwaqqas

Sec7

This repository contains several independent C++ command-line programs demonstrating fundamental programming concepts such as control flow, operators, and a basic student record management system; it does not implement a server.

Setup Requirements

  • ⚠️Requires a C++ compiler (e.g., GCC/G++).
  • ⚠️The use of Variable Length Arrays (VLAs) in 'studentrecord.cpp' is a C99 feature and a GNU C++ extension, which may not compile with all standard C++ compilers (e.g., MSVC without specific flags). For broader compatibility, `std::vector` should be used instead.
Verified SafeView Analysis
The primary 'studentrecord.cpp' file uses Variable Length Arrays (VLAs) (`Student s[num];`) based on user input, which is a non-standard C++ feature and a GNU C++ extension. If 'num' is excessively large, this can lead to stack overflow, potentially causing a program crash. The manual array manipulation for 'deletion' is inefficient and error-prone. However, as these are simple command-line applications without network exposure, hardcoded secrets, or dynamic code execution ('eval'-like functions), the overall security risk is low for their intended educational use.
Updated: 2026-01-19GitHub→
0
0
Low Cost
huberp icon
Sec4

MCP server with OAuth 2.1 authentication for secure GitHub API access, designed to run locally in MCP hosts like Visual Studio Code.

Setup Requirements

  • ⚠️Requires GitHub OAuth App credentials (OAUTH_CLIENT_ID, OAUTH_CLIENT_SECRET) which involve a multi-step manual setup on GitHub.
  • ⚠️Requires Python 3.12+.
  • ⚠️Docker is recommended for deployment, adding a dependency for development environment setup.
Review RequiredView Analysis
Critical vulnerabilities exist in the current implementation that make it unsafe for production or external deployment: In-memory OAuth session storage (lost on restart, not scalable, susceptible to session fixation/hijacking); lack of robust server-side token expiry validation and automated refresh; and missing rate limiting. The default HTTP transport is unencrypted, exposing OAuth tokens and making it vulnerable to Man-in-the-Middle attacks if not behind a secure HTTPS reverse proxy (explicitly warned in documentation). Additionally, input validation is not comprehensive for all tool parameters, and audit logging is missing. Generic exception catching is also present in some areas.
Updated: 2025-12-15GitHub→
0
0
High Cost
boxcalfdevelopmentallearning748 icon

OpenScout

by boxcalfdevelopmentallearning748

Sec8

OpenScout is a Perplexity-inspired answer engine that provides well-researched and cited responses to user questions using web search, retrieval, and LLM synthesis.

Setup Requirements

  • ⚠️Requires multiple external API keys (at least one LLM, Tavily, and OpenAI for embeddings).
  • ⚠️Uses local FAISS index and SQLite database for storage, consuming disk space and memory during operation.
Verified SafeView Analysis
The application handles API keys by loading them from `.env` or allowing user input in the UI, which is a good practice. It explicitly states keys are not stored or logged. SQLite is used with parameterized queries, preventing SQL injection. While `unsafe_allow_html=True` is used in Streamlit, it appears in contexts where content is either static or derived from sanitized text (e.g., `trafilatura` output), reducing immediate XSS risks. The fetching of external web content via `httpx` and `trafilatura` introduces inherent risks associated with processing arbitrary external data, but standard libraries are used for parsing.
Updated: 2026-01-19GitHub→
0
0
Low Cost
fast-ish icon
Sec9

A template for building Model Context Protocol (MCP) servers, enabling AI assistants to interact with external tools, data, and reusable prompts in a structured way.

Setup Requirements

  • ⚠️Requires manual configuration of `~/.config/Claude/claude_desktop_config.json` to connect to Claude Desktop.
  • ⚠️Specific package managers are recommended for each language (pnpm for TypeScript, uv for Python), which may require prior installation.
  • ⚠️File system access capabilities depend on setting the `MCP_ALLOWED_DIR` environment variable to a secure, allowed directory.
  • ⚠️Requires either Node.js (v18, 20, or 22) or Python (v3.11, 3.12, or 3.13) depending on the chosen language.
Verified SafeView Analysis
The server template explicitly incorporates strong security principles outlined in its 'SECURITY.md'. Key measures include: strict input validation (Zod/Pydantic), robust path traversal prevention for file system access via `MCP_ALLOWED_DIR` environment variable, comprehensive URL validation (blocking private networks, enforcing HTTPS) for HTTP client tools, guidance on parameterized queries for database interactions to prevent SQL injection, and a focus on secure error handling. Authentication via OAuth 2.1 is advised for HTTP transport. No 'eval' or malicious patterns were found, and it avoids hardcoded secrets by directing users to use environment variables.
Updated: 2025-12-14GitHub→
0
0
Medium Cost
mikebmac86 icon

pviz-mcp-server

by mikebmac86

Sec9

Exposes polyglot dependency analysis capabilities (circular dependencies, metrics, architecture comparison) for Python, TypeScript, JavaScript, Java, and Go projects to LLMs via the Model Context Protocol.

Setup Requirements

  • ⚠️Requires a pvizgenerator.com account with tokens for API usage (paid service).
  • ⚠️Requires Python 3.8+.
  • ⚠️A JWT token must be configured via environment variable (PVIZ_JWT_TOKEN) or file (PVIZ_JWT_TOKEN_FILE).
Verified SafeView Analysis
The server securely handles JWT tokens (from environment or file) and GitHub Personal Access Tokens (PATs). PATs are transmitted over HTTPS to the backend, used only for repository cloning, and explicitly stated to be never stored or logged by the server. There is no 'eval', obfuscation, or direct shell command execution with unsanitized user input in the provided code. Communication to the backend API defaults to HTTPS. The primary security responsibility for repository processing and data persistence lies with the pviz backend API, whose practices are well-documented.
Updated: 2026-01-19GitHub→
0
0
Medium Cost
swethasalunke-tech icon

postgres-mcp-server

by swethasalunke-tech

Sec9

Provides an AWS Model Context Protocol (MCP) server for secure, controlled interaction with PostgreSQL databases via the RDS Data API, supporting SQL query execution and schema retrieval.

Setup Requirements

  • ⚠️Requires an AWS RDS PostgreSQL instance and an AWS Secrets Manager secret for database credentials.
  • ⚠️Requires specific AWS IAM permissions for rds-data actions on the specified RDS cluster and secret.
  • ⚠️Needs resource_arn, secret_arn, database, region, and readonly flag to be provided as command-line arguments.
Verified SafeView Analysis
The server includes robust, explicit checks to prevent SQL injection and detect mutating SQL queries, especially when configured for read-only access. It leverages AWS IAM roles and Secrets Manager via boto3 for secure credential management, avoiding hardcoded secrets. Parameterized queries are utilized where appropriate.
Updated: 2025-11-20GitHub→
0
0
Medium Cost
rdwj icon
Sec8

Provides real-time and historical US weather data, forecasts, and generates weather reports using MCP (Model Context Protocol) for consumption by AI/LLMs.

Setup Requirements

  • ⚠️Requires `NOAA_CDO_TOKEN` environment variable (obtain free API key from NOAA) for historical weather data.
  • ⚠️Requires OpenShift CLI (`oc`) for deployment to OpenShift clusters.
  • ⚠️Python 3.11+ is required.
Verified SafeView Analysis
The server uses environment variables for sensitive API keys (e.g., NOAA_CDO_TOKEN) and JWT secrets, which is good practice. Dynamic loading of tools, resources, and prompts is a core feature of the FastMCP framework; this is safe assuming the source code is trusted and not modifiable by untrusted users at runtime. JWT authentication is optional, and if not configured, the HTTP endpoint will be open. CORS configuration (MCP_HTTP_ALLOWED_ORIGINS) is mentioned in the architecture but not explicitly configured in the `mcp.run` call, which might require additional FastMCP or reverse proxy configuration for production. No direct `eval()` or `exec()` usage was found that would allow arbitrary code execution through user input.
Updated: 2025-12-05GitHub→
PreviousPage 724 of 760Next