mcp -security-standard
MCP Server Security Standard (MSSS): an open, testable security control standard for certifying MCP servers, with levels, evidence requirements, and reporting schemas.
claude mcp add --transport stdio mcp-security-standard-mcp-server-security-standard docker run -i mcp-server-security-standard:latest
How to use
The MCP Server Security Standard (MSSS) implements a structured, risk-based approach to enabling AI tools and prompts while enforcing strong security controls. MSSS exposes a configurable MCP server that coordinates tool usage, input validation, access control, and auditing across deployment profiles. The server is designed to be deployed using containerization (Docker in this guide) to ensure reproducible environments and integrated with reporting schemas for machine-readable verification. Tools, profiles, and controls are aligned with the MSSS 24 security controls across 8 domains, supporting automated verification and threat modeling as your AI deployment scales.
Once running, you can interact with the MCP server to validate tool invocations, enforce arguments and input constraints, perform authorization checks, and generate audit logs. The included reporting schemas enable you to produce JSON-based evidence of compliance with the selected level (L1–L4) and deployment profile. Use the deployment profiles to tailor the security posture to your environment, then assess and map controls to your real-world usage to maintain continuous assurance.
How to install
Prerequisites:
- Docker installed and running on your host
- Git installed
- Access to a terminal/CLI
Install and run MSSS using Docker:
-
Pull the latest MSSS image (or build locally if you have a Dockerfile): docker pull mcp-server-security-standard:latest
-
Run the MCP server container: docker run -d --name mcp-server-security-standard -p 8080:80 mcp-server-security-standard:latest
-
Verify the server is up: docker ps curl http://localhost:8080/health
Optional: clone and run from source (if you prefer developing against the repository):
- git clone https://github.com/mcp-security-standard/mcp-server-security-standard.git
- cd mcp-server-security-standard
- Follow any repository-specific install steps in the README (e.g., install dependencies) and run the server as described there.
Notes:
- Adjust port mappings as needed to fit your environment and firewall rules.
- If you use TLS/HTTPS in front of the MCP server, ensure certificates are mounted and the reverse proxy is configured accordingly.
Additional notes
Tips and reminders:
- Review the deployment profiles in v0.1/standard/profiles.md to choose the appropriate security posture for your environment.
- Ensure you enable the appropriate Level (L1–L4) based on your data sensitivity and risk tolerance; higher levels provide stronger protections but may add overhead.
- Enable comprehensive audit logging (MSSS LOG controls) to support incident response and compliance reporting.
- Regularly verify that input validation (MCP-INPUT-01, MCP-INPUT-02) and command execution controls (MCP-EXEC-01, MCP-EXEC-02) are enforced in your runtime environment.
- If using Docker, keep image provenance in mind and pin to a specific version tag to prevent unexpected updates; periodically audit the image for vulnerabilities.
- For testing, use the JSON reporting schemas to generate machine-readable attestations of compliance for internal audits or external assessments.
Related MCP Servers
mcp-for-beginners
This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable, and secure AI workflows from session setup to service orchestration.
ENScan_GO
一款基于各大企业信息API的工具,解决在遇到的各种针对国内企业信息收集难题。一键收集控股公司ICP备案、APP、小程序、微信公众号等信息聚合导出。支持MCP接入
cloudsword
一款帮助云租户发现和测试云上风险、增强云上防护能力的综合性开源工具
mcp-for-security
MCP for Security: A collection of Model Context Protocol servers for popular security tools like SQLMap, FFUF, NMAP, Masscan and more. Integrate security testing and penetration testing into AI workflows.
mcp-security-hub
A growing collection of MCP servers bringing offensive security tools to AI assistants. Nmap, Ghidra, Nuclei, SQLMap, Hashcat and more.
pentesting s-checklist
A practical, community-driven checklist for pentesting MCP servers. Covers traffic analysis, tool-call behavior, namespace abuse, auth flows, and remote server risks. Maintained by Appsecco and licensed for remixing.