Back to Feed

Introducing The Backslash MCP Server Security Hub

-

June 16, 2025

Amit Bismut

June 16, 2025

Today we’re very excited to introduce a new, free resource for vibe coders, developers and security teams - the Backslash MCP Server Security Hub. It is the first dynamically maintained, searchable central database of MCP servers that rates their risk and security posture based on a variety of factors including vulnerabilities and their severity, exposure to attack vectors, and provenance factors.

Additionally, for a limited time, we’re offering a free self-assessment tool and dashboard of your MCP server implementation risks. The link to register is available on the MCP Server Security Hub website.

Why We Built MCP Server Security Hub

As organizations rapidly adopt AI agents, copilots, and autonomous tools, a new category of backend infrastructure has emerged: Model Context Protocol (MCP) servers. These servers enable secure, contextual orchestration of AI behavior, and they are being adopted and used very quickly across all aspects of software development, applications, and IT. However, their rise introduces a new and largely unprotected attack surface. We highlighted the top 10 sks MCP servers present specifically when used in IDEs in a previous blog. 

Security teams are increasingly under pressure to assess and approve the use of MCPs across their organizations. This need is especially acute as developers adopt AI-enhanced IDEs such as Cursor, Windsurf, and Copilot, leveraging MCPs to accelerate development workflows. We described how MCP server security presents new challenges to security teams, who need to quickly assess them (assuming they are even aware of their use). 

Today, there are more than 15,000 MCP servers available publicly. There are already several directories and websites cataloging MCPs (such as mcp.so, Awesome MCP Servers, PulseMCP, MCP Market, Docker’s MCP list, a list within the Github MCP project, and more). Most MCP servers are not published by verified sources and lack proper security vetting, even as they serve as gateways to sensitive services, tools, and resources. Until now there’s been no public resource that rates MCP servers on their security posture - and we’re here to fix that.

The Backslash MCP Server Security Hub is the first security-focused database built specifically for this evolving layer of infrastructure.

What You Can Do with MCP Server Security Hub

The hub (for short) is free to use and does not require logging in. It uses a familiar directory interface, providing quick access to the most popular and most risky MCPs. Naturally you can search MCP servers by name. The database already includes thousands of MCPs and we are continuously adding to it.

  1. Search and Score: Look up any MCP server to instantly view its security score and quickly determine whether it’s safe to use.
  2. Identify Risks: Review detailed security risks associated with each MCP server. This includes looking for malicious patterns, and weaknesses in the code such as unauthorized network exposure or excessive permissions. 
  3. Detect attack vector: See whether a specific MCP server has a Local or Network attack vector.
  4. Get MCP details: Check whether an MCP server originates from a verified GitHub publisher and review its documentation.
  5. Coming Soon… additional security-focused features and use cases are on the way.

Things are moving extremely fast – and security teams are “flying blind” trying to ensure that risk is being managed. Assessing MCP server risks is a good starting point.

👉 Explore the MCP Server Security Hub now