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
High Cost
GizAI icon
Sec3

Provides a Claude Code MCP server for browser automation and full local PC control, enabling AI agents to interact with web browsers and the local operating system.

Setup Requirements

  • ⚠️Requires Python 3.10+.
  • ⚠️Requires a locally installed Chrome, Chromium, or Edge browser for browser tools.
  • ⚠️Full PC control mode (`--local`) operates with no sandbox, granting unrestricted system access.
Review RequiredView Analysis
The server's core functionality (`shell()` and `python_exec()`) allows for arbitrary code execution and full system access, including file system, process management, and screenshots. This is explicitly stated as 'NO restrictions - use responsibly' by the developers. Browser control also includes `eval()` for JavaScript. While remote access requires token authentication by default and local PC control needs explicit `--local` flag, the inherent power of this tool presents a maximal security risk if the server or its authentication token is compromised, or if used irresponsibly. It is designed for maximum control, not maximum sandbox security.
Updated: 2025-12-08GitHub→
0
0
High Cost
Liquescent-Development icon

mcprobe

by Liquescent-Development

Sec6

A conversational testing framework for MCP (Model Context Protocol) servers, validating that LLM agents can correctly answer real-world questions using synthetic users and LLM judges.

Setup Requirements

  • ⚠️Requires Python 3.11 or higher.
  • ⚠️Requires an LLM provider: either `ollama serve` running locally with a model pulled (e.g., `llama3.2`), or an `OPENAI_API_KEY` for a paid cloud service.
  • ⚠️For testing Gemini ADK agents, requires `mcprobe[adk]` installation, `GOOGLE_API_KEY`, and a separate MCP server to be running and accessible via `MCP_URL` and `MCP_TOKEN` environment variables.
Verified SafeView Analysis
The `mcprobe generate-scenarios` command executes arbitrary user-provided commands (e.g., `npx @example/weather-mcp`) via `subprocess.run` to connect to an MCP server. While intended for developers to launch their *own* trusted local servers for testing, this feature represents a significant security risk if the `server` argument were ever exposed to untrusted input. Otherwise, the project follows good security practices, such as using `SecretStr` for API keys in configurations and relying on environment variables.
Updated: 2026-01-19GitHub→
0
0
Low Cost
Letitia-May icon

mcp-recipe-collector

by Letitia-May

Sec6

A Model Context Protocol (MCP) server that acts as a proxy for AI assistants to search, retrieve, and add recipes to a local recipe API.

Setup Requirements

  • ⚠️Requires a separate, locally running recipe API server at `http://127.0.0.1:8080` to function, which is a critical external dependency.
  • ⚠️The direct execution command `node index.ts` for a TypeScript file (`index.ts`) may require a global `tsx` installation or a prior TypeScript compilation step, which is not explicitly detailed in the 'Usage' instructions for running the MCP server.
Verified SafeView Analysis
The MCP server makes HTTP requests to a local recipe API (http://127.0.0.1:8080). For the 'searchRecipes' tool, the user-provided 'term' parameter is directly interpolated into the URL query string. If the underlying recipe API does not implement robust input sanitization for this 'query' parameter, it could be vulnerable to injection attacks (e.g., SQL injection, command injection) if a sophisticated attacker compromises the client interacting with this MCP server. The 'id' parameter for 'getRecipeById' is validated as a number using Zod, mitigating direct injection risk for that endpoint. No obvious hardcoded secrets or malicious patterns were found in the provided source code.
Updated: 2026-01-12GitHub→
0
0
Medium Cost
dnys1 icon

mcp

by dnys1

Sec8

Provides Model Context Protocol (MCP) servers for AI assistants, enabling semantic search over documentation and multi-project TODO management.

Setup Requirements

  • ⚠️Requires OpenAI API Key (Paid) for mcp-docs server (embeddings and description generation).
  • ⚠️Requires Firecrawl API Key (Paid, conditional) for mcp-docs server if using Firecrawl web scraping for ingestion.
  • ⚠️Requires Bun runtime.
Verified SafeView Analysis
The project uses `execSync` for Git commands and MCP CLI configuration, which can be risky if the current working directory (`cwd`) or arguments are maliciously controlled, though inputs here are largely internal or user-configured. The documentation server (`mcp-docs`) fetches content from user-provided URLs (`llms.txt` or Firecrawl), which could expose to Server-Side Request Forgery (SSRF) if URLs point to internal network resources and proper mitigations are not in place. API keys for OpenAI and Firecrawl are handled via environment variables, not hardcoded. The database is local SQLite.
Updated: 2025-12-13GitHub→
0
0
Medium Cost
kfzzzzzz icon

gaode-mcp-server

by kfzzzzzz

Sec5

This server likely provides an API to interact with or process data related to Gaode Maps or a mobile content platform.

Setup Requirements

  • ⚠️Requires Gaode API Key (potentially paid service)
  • ⚠️Requires a database (e.g., MySQL, MongoDB)
Review RequiredView Analysis
A thorough security audit is impossible as no source code was provided for analysis. The score reflects an inability to verify safety rather than identified vulnerabilities. Users should exercise extreme caution as hardcoded secrets, dangerous patterns like 'eval', or malicious code cannot be ruled out without actual code review.
Updated: 2025-12-13GitHub→
0
0
Low Cost
Sec9

Demonstrates Gilhari's INLINE mapping feature for performance optimization in RESTful ORM by storing child object attributes in the parent's database table row.

Setup Requirements

  • ⚠️Docker is required to build and run the microservice.
  • ⚠️Java Development Kit (JDK 1.8+) is required for recompilation if source code is modified.
  • ⚠️The Gilhari SDK is required for development and customization.
Verified SafeView Analysis
The server runs as a local Docker container and exposes a REST API on port 80/8081. As an example, it's designed for local execution and is safe in that context. No apparent use of 'eval' or obfuscation. Standard network exposure for a microservice.
Updated: 2025-11-18GitHub→
0
0
Low Cost

studio-mcp

by studio-mcp

Sec10

Provides a convenient alias for the `@studio-mcp/studio` CLI, simplifying command execution.

Setup Requirements

  • ⚠️Node.js and npm/npx runtime required
Verified SafeView Analysis
This package acts as a thin alias/wrapper for the `@studio-mcp/studio` CLI. It does not contain complex logic, 'eval' usage, or obfuscation. Direct security risks from this specific package are minimal. Security considerations would primarily depend on the underlying `@studio-mcp/studio` package it invokes.
Updated: 2025-11-17GitHub→
0
0
Low Cost

fragrance-mcp

by lflor120

Sec5

A server for managing and processing fragrance-related data.

Review RequiredView Analysis
Insufficient information to perform a detailed security audit due to lack of code beyond .gitignore and README.md. No known risks, but no verification of safety possible.
Updated: 2025-11-22GitHub→
0
0
High Cost
albertjoseph0 icon
Sec8

Provides an authenticated Model Context Protocol (MCP) server that exposes proprietary data sources, specifically travel-industry expert-call transcripts from an OpenAI Vector Store and structured airfare trend data from local files, to ChatGPT or other MCP-capable clients.

Setup Requirements

  • ⚠️Requires Python 3.10+
  • ⚠️Requires an Auth0 tenant (or any OAuth 2.1 provider with OIDC discovery) for authentication setup
  • ⚠️Requires an OpenAI API Key (Paid Service) for vector store operations
  • ⚠️Requires `VECTOR_STORE_ID` to be populated by running an initial script (e.g., `python scripts/upload_expert_calls_to_vector_store.py`)
  • ⚠️RESOURCE_SERVER_URL needs to be updated to a public URL for external clients like ChatGPT
Verified SafeView Analysis
The server uses Auth0 for OAuth 2.1 bearer token authentication with JWT validation via JWKS, which is a standard and robust approach. Environment variables are correctly utilized for sensitive information like API keys and issuer URLs, preventing hardcoding. The Python `JWTVerifier` explicitly notes that it 'does not contain entitlements' and 'you must add your own entitlements,' indicating that fine-grained authorization logic beyond basic token validity needs to be implemented by the developer. The `SimpleTokenVerifier` is present but not used in the main application. Overall, it provides a secure foundation, but requires developer attention for full authorization implementation.
Updated: 2025-12-01GitHub→
0
0
High Cost
praneet1503 icon

astrx-mcp-server

by praneet1503

Sec9

A high-performance RAG system providing AI-enriched knowledge and semantic search capabilities for a large animal dataset.

Setup Requirements

  • ⚠️Requires API keys for SambaNova, Anthropic Claude, Google Gemini, and Blaxel (or relies on limited demo keys).
  • ⚠️Requires Modal CLI installed and authenticated (`modal setup`) for AI enrichment and semantic search embeddings.
  • ⚠️Initial data setup (`data/animals.json`, external CSVs) is required, although `scripts/download_and_merge.py` can automate parts of this.
Verified SafeView Analysis
API keys are handled via environment variables, `.env` files, and Modal secrets, or temporarily stored in session memory for Gradio. Network calls to external LLM APIs use `httpx` with timeouts. Web scraping with Playwright is isolated within Modal functions. No direct `eval` or `subprocess` execution from user input is observed. The data loading is from controlled local files or specific external URLs. Standard prompt injection risks for LLM interactions exist but do not compromise the underlying system.
Updated: 2025-11-30GitHub→
0
0
Medium Cost

kali-mcp

by JesseEikeland

Sec8

Provides a Dockerized Kali Linux environment with a Python server (MCP server) to programmatically access and orchestrate various penetration testing tools.

Setup Requirements

  • ⚠️Docker and Docker Compose are required.
  • ⚠️At least 4GB of free disk space is needed.
  • ⚠️Initial Docker container build takes 10-15 minutes to download and install all tools.
Verified SafeView Analysis
The project wraps powerful offensive security tools. While the README emphasizes robust security features for the server's implementation (non-root user, input sanitization, command injection prevention, timeouts, resource limits, minimal capabilities), the inherent nature of the tools means misuse can lead to severe legal consequences. Extensive legal disclaimers are provided. Users must exercise extreme caution and only use it on authorized systems.
Updated: 2025-11-17GitHub→
0
0
Low Cost
myvfc icon
Sec8

Provides college football player statistics from the CFBD API through an MCP (Multi-tool Coordination Protocol) endpoint.

Setup Requirements

  • ⚠️Requires a free CFBD API Key (needs sign-up at collegefootballdata.com).
  • ⚠️An `MCP_API_KEY` is recommended for securing the `/mcp` endpoint.
  • ⚠️The provided JavaScript server files (`server.js`, `cfbd-stats-server.js`, etc.) only implement the `get_player_stats` tool, despite documentation (CFBD_COMPLETE_PROMPT.md, CFBD_README.md) describing 8 comprehensive tools.
Verified SafeView Analysis
The server correctly uses environment variables for `CFBD_API_KEY` and `MCP_API_KEY`, preventing hardcoded secrets. It implements optional bearer token authentication for the `/mcp` endpoint. Input validation is minimal for `team` and `year` parameters, relying on CFBD API for handling, which is generally acceptable for public APIs. Extensive logging in some variants could expose request details, but it's not a critical vulnerability. CORS is implicitly open for all origins as `cors()` is used without specific origin restrictions in some included files, which is common for tool servers but could be tightened if needed. The primary server.js does not explicitly configure cors, but express does not restrict it by default. The server does not use `eval` or other highly dangerous patterns.
Updated: 2025-11-20GitHub→
PreviousPage 726 of 760Next