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)

33
3
High Cost
jpmorgan-payments icon

pdp-mcp

by jpmorgan-payments

Sec9

Provides an AI agent with tools to search, read, and find related documentation for JPMorgan Chase Payments Developer Portal APIs.

Setup Requirements

  • ⚠️Requires Python 3.10 or newer.
  • ⚠️Requires `uv` or `pip` for package management, and `git` for cloning.
  • ⚠️May require `HTTP_PROXY` or `HTTPS_PROXY` environment variables to be configured if operating behind a corporate network proxy.
Verified SafeView Analysis
The MCP server codebase (Python files) generally follows good security practices: URL validation is enforced for `read_documentation` and `related` tools to prevent Server-Side Request Forgery (SSRF) by ensuring URLs are from the expected JPMC domain. `httpx` is used for network requests, and environment variables are used for proxy configuration. HTML parsing utilizes `BeautifulSoup`, which is robust against malformed HTML. No direct use of `eval`, `os.system`, or `subprocess.run(shell=True)` was found in the server's Python code. No hardcoded secrets were identified. The `metrics/merge.sh` script, while present in the repository, is located in a separate `metrics` directory and is not part of the core MCP server's runtime logic, hence its file handling operations are not assessed as a security risk for the server itself.
Updated: 2026-01-19GitHub→
33
3
Medium Cost
cloudy-sfu icon
Sec5

Collects data (posts, search results, post details) from the Xiaohongshu social media platform.

Setup Requirements

  • ⚠️Requires a Xiaohongshu account and manual cookie extraction using a browser extension (J2TEAM Cookies or alternative).
  • ⚠️Requires Python 3.13.
  • ⚠️Requires Cherry Studio ≥ 1.7.8 running on Windows ≥ 10.
Verified SafeView Analysis
The server requires the user to manually extract and provide their Xiaohongshu login cookies, which contain sensitive session and authentication data. These cookies are stored locally in `raw/cookies.csv`. If the system running this MCP server is compromised, the user's Xiaohongshu account session could be exposed and misused. While the code itself does not contain explicit 'eval' or malicious patterns, the reliance on storing user session cookies locally represents a significant security risk to the user's account if the local environment is not adequately secured.
Updated: 2026-01-17GitHub→
33
2
High Cost
tobs-code icon

a-mem-mcp-server

by tobs-code

Sec8

An agentic memory system for LLM agents that stores, links, evolves, and retrieves information based on the Zettelkasten principle, enhanced with autonomous maintenance enzymes and deep web research capabilities.

Setup Requirements

  • ⚠️Requires either a local Ollama server running with 'qwen3:4b' and 'nomic-embed-text:latest' models, or an OpenRouter API key (paid service).
  • ⚠️The Researcher Agent (if enabled) may require additional setup: Google Search API Key and Search Engine ID, a local Docker-based Jina Reader instance, or an Unstructured PDF extraction API endpoint (or local library with heavy dependencies).
  • ⚠️The 'falkordb' graph backend, especially on Windows, requires an external Redis server running with the FalkorDB module and additional Python package installations.
Verified SafeView Analysis
The system externalizes API keys and configurations via .env files. It implements input validation for MCP tool parameters, reducing common injection risks. The optional HTTP server for graph visualization is read-only. External network calls (Ollama, OpenRouter, Google Search, Jina Reader, Unstructured) are handled using the 'requests' library, relying on standard API security. However, the Researcher Agent's web content fetching and processing of external data sources introduces potential attack surfaces (e.g., if URLs are not perfectly sanitized before fetching, or if fetched content could lead to processing vulnerabilities, though current implementations use robust libraries like Jina Reader and Unstructured). Local file I/O for data storage (graph, chroma, events) requires the host system to have appropriate file permissions.
Updated: 2026-01-10GitHub→
33
3
Medium Cost
hu-qi icon

starReport

by hu-qi

Sec9

starReport is a Node.js tool for automatically generating and managing GitHub repository activity reports (stars, commits, issues), supporting MCP protocol, integrating large model AI analysis, and pushing reports to Feishu groups.

Setup Requirements

  • ⚠️Requires a GitHub Personal Access Token (`GITHUB_TOKEN`) with appropriate repository read permissions.
  • ⚠️Requires a Feishu group bot Webhook URL (`FEISHU_WEBHOOK`) for notifications.
  • ⚠️Requires an API Key (`API_KEY` or `OPENAI_API_KEY`) for a compatible Large Language Model (e.g., OpenAI, Zhipu AI), which may incur usage costs.
  • ⚠️The list of GitHub repositories to monitor (`REPORT_REPOS`) must be explicitly configured.
Verified SafeView Analysis
The server loads sensitive credentials (GitHub Token, Feishu Webhook, AI API Key) from environment variables, which is a good practice. Data is stored locally in a configurable `DATA_FILE`, with a fallback to `/tmp` if the primary path is read-only; ensure `DATA_FILE` points to a secure, writable location to prevent data loss and unauthorized access. User input from Feishu webhooks is passed to the LLM for analysis; while structured prompts aim to mitigate risks, the security relies on the robustness of the integrated LLM against potential prompt injection scenarios. No 'eval' or obvious obfuscation found.
Updated: 2026-01-19GitHub→
33
3
Medium Cost

OpenFoodFacts-MCP

by JagjeevanAK

Sec9

Provides contextual information and specialized tools for Open Food Facts data and development tasks to AI-powered development tools using the Model Context Protocol.

Setup Requirements

  • ⚠️Requires Node.js (v16.x or higher) and npm/yarn for installation and running.
  • ⚠️Requires 'Model Context Protocol' VS Code extension for seamless VS Code integration.
  • ⚠️GitHub API tools in Developer Mode are subject to unauthenticated GitHub API rate limits, which may impact heavy usage.
Verified SafeView Analysis
The server operates locally or internally, interacting with external APIs (Open Food Facts and GitHub). It does not appear to use 'eval' or obfuscation. Network risks are minimal given its intended local deployment; however, for developer mode, GitHub API rate limits could be a functional issue, not a security flaw. Standard Node.js security practices should be followed for deployment.
Updated: 2025-11-17GitHub→
33
1
Medium Cost
julienkalamon icon

ign-apicarto-mcp-server

by julienkalamon

Sec9

This server provides access to French geographical data from the IGN API Carto, enabling natural language queries for cadastral, agricultural, environmental, urban planning, and administrative information through the Model Context Protocol (MCP).

Setup Requirements

  • ⚠️Requires Node.js 16+ and npm to be installed.
  • ⚠️An IGN API Key (free to obtain) is required for the 'ign_get_aoc_viticoles' tool and is recommended for full functionality, to be set in a '.env.local' file or as an environment variable.
  • ⚠️Client configuration (e.g., Claude Desktop) typically requires the absolute path to the 'dist/index.js' file.
Verified SafeView Analysis
The server uses Zod for input validation, which helps prevent malformed inputs and potential injection. It interacts with external IGN APIs via standard HTTP requests and does not use 'eval' or other direct code execution methods. The IGN_API_KEY is handled as an environment variable or parameter and is not hardcoded. Responses are truncated by a character limit, mitigating large data exfiltration or resource exhaustion through the proxy itself. Security largely relies on the underlying IGN API's robustness.
Updated: 2025-12-02GitHub→
33
3
Low Cost
konflux-ci icon

konflux-devlake-mcp

by konflux-ci

Sec9

The Konflux DevLake MCP Server provides a natural language interface for AI assistants to query Konflux DevLake databases, enabling intuitive data analysis and reporting without needing to write SQL.

Setup Requirements

  • ⚠️Requires an existing MySQL database instance (DevLake compatible schema expected)
  • ⚠️Requires Python 3.11+
  • ⚠️An external LLM (e.g., Gemini, OpenAI) API Key is needed for natural language querying functionality (though not for the server itself to run).
Verified SafeView Analysis
The server demonstrates a strong focus on security, particularly against SQL injection, with multiple layers of validation (regex, keyword blacklisting, balanced parentheses check, SELECT-only policy) and explicit warnings for the `execute_query` tool. Sensitive data masking is implemented. Hardcoded secrets are present only in development/test `docker-compose.yml` and are explicitly excluded from Kubernetes deployment via secrets. Connection timeouts are also configured for robustness. The use of a 'raw SQL query tool', even with extensive safeguards, is an inherent risk that requires careful AI orchestration.
Updated: 2026-01-18GitHub→
33
2
Medium Cost

mcp-c

by servoagents

Sec7

A C/C++ based server likely focused on high-performance networking, potentially for a game-related application like a Minecraft protocol implementation.

Setup Requirements

  • ⚠️Requires a C++ compiler (e.g., GCC or Clang) and CMake for building the server.
  • ⚠️Python and its dependencies listed in `requirements.txt` are likely needed for build scripts, development tools, or testing.
Verified SafeView Analysis
As a C/C++ server, it operates at a low level, requiring careful memory and network handling to prevent vulnerabilities such as buffer overflows or denial-of-service attacks. No 'eval' or explicit obfuscation mechanisms are indicated by the file list, but a full audit would necessitate code review.
Updated: 2025-11-18GitHub→
33
3
Medium Cost
sultannaufal icon

puppeteer-mcp-server

by sultannaufal

Sec7

Automating browser interactions and web scraping via an MCP server.

Setup Requirements

  • ⚠️Requires `API_KEY` environment variable for authentication.
  • ⚠️Requires Docker for containerized deployment, or manual installation of Node.js (v18+) and Chromium with its system dependencies for traditional server deployment.
  • ⚠️The `puppeteer_evaluate` tool allows executing arbitrary JavaScript, which carries inherent security risks even with implemented validation, and should be used with caution.
Verified SafeView Analysis
The `puppeteer_evaluate` tool allows executing arbitrary JavaScript code in the browser, which is an inherent security risk, even with input validation. The `allowDangerous` flag in `puppeteer_navigate` can also expose the browser to less secure configurations if explicitly enabled. However, the project implements robust security measures including API key authentication, URL, script, and selector safety validations, Helmet.js, CORS, rate limiting, and runs Puppeteer in a sandboxed, non-root Docker environment. Binary image serving is not authenticated but relies on temporary, UUID-based URLs and automatic cleanup.
Updated: 2025-11-23GitHub→
33
3
Medium Cost
vespo92 icon

QBO-MCP-TS

by vespo92

Sec9

Integrate with QuickBooks Online to provide modular accounting automation, financial reporting, and real-time updates via an MCP server.

Setup Requirements

  • ⚠️Requires a QuickBooks Online Developer Account with configured OAuth credentials (CLIENT_ID, CLIENT_SECRET, COMPANY_ID, REFRESH_TOKEN).
  • ⚠️Requires Node.js 18+ to run.
  • ⚠️SSE transport is temporarily disabled for migration and will not function as expected (as stated in `src/index.ts`).
  • ⚠️Requires WEBHOOK_VERIFIER_TOKEN if QuickBooks webhooks are to be utilized for real-time updates.
Verified SafeView Analysis
Relies on environment variables for sensitive data (QuickBooks OAuth credentials, webhook verifier token). Implements HMAC-SHA256 signature verification for inbound webhooks and includes a basic replay attack protection mechanism. Logging is sanitized to prevent sensitive data exposure. No 'eval' or obvious obfuscation detected.
Updated: 2025-11-24GitHub→
33
1
Low Cost
ahnafyy icon
Sec8

Provides a Model Context Protocol (MCP) server template with Express.js for building AI-powered tools and resources.

Setup Requirements

  • ⚠️Requires Node.js >= 18.0.0.
  • ⚠️MCP Inspector is an optional, external debugging tool (`mcp-inspector`) that needs to be installed or run via `npx` separately.
Verified SafeView Analysis
The core server template itself does not use 'eval' or direct 'exec' functions. Configuration is loaded from environment variables, which is a good practice. The `requestLogger` middleware logs `req.body` and `req.query` at debug level, which could potentially expose sensitive data if debug logging is enabled in a production environment without careful consideration. The MCP request handling includes a `try-catch` block that prevents server crashes for individual malformed requests, returning a 500 Internal Server Error. The scaffolding tool `create-mcp-express` uses `child_process.execSync` for package installation and fetching git user config, which is generally acceptable for a local CLI tool but not part of the runtime server.
Updated: 2025-11-23GitHub→
33
3
Low Cost

tigris-mcp-server

by tigrisdata

Sec8

Integrates AI agents with Tigris S3-compatible object storage for managing buckets and objects, enabling AI-driven data interactions.

Setup Requirements

  • ⚠️Requires a Tigris account and S3-compatible access keys (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY).
  • ⚠️Requires either Docker Engine or Node.js/NPM to be installed locally.
  • ⚠️Manual configuration of the MCP server within the AI client (e.g., VS Code, Claude Desktop) is needed, including setting environment variables for credentials.
Verified SafeView Analysis
The server requires AWS S3-compatible credentials (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY) with access to your Tigris account. Running via 'npx' carries inherent risks associated with executing third-party code, though it's a common practice for CLI tools. The recommended Docker installation provides better sandboxing, explicitly limiting file system access via bind mounts to a specified user directory. Users should ensure their environment variables containing credentials are secure.
Updated: 2025-11-17GitHub→
PreviousPage 171 of 760Next