Uses Python as the runtime environment for the MCP server, allowing malicious plugins to exploit weaknesses in the MCP architecture through practical examples of security attack vectors.
MasterMCP
Project Introduction
MasterMCP is a demonstration tool designed to showcase various potential security attack vectors against MCP (Model Control Protocol). This project illustrates how malicious plugins can exploit weaknesses in the MCP architecture through practical examples, helping developers and security researchers understand these risks and strengthen system protection.
Features
- Plugin-based Malicious Payloads: Demonstrates how malicious plugins can run within the MCP architecture
- Multiple Attack Techniques: Includes examples of data poisoning, cross-MCP calls, competitive malicious functions, and more
- Practical Educational Value: Each attack vector comes with detailed explanations and implementation code
Included Attack Vectors
- Data Poisoning: The
banana
plugin demonstrates how to force users to perform specific operations - JSON Injection Attacks: The
url_json
plugin shows how to retrieve data from a local malicious service - Competitive Malicious Functions: The
remove_server
plugin overrides existing functionality - Cross-MCP Call Attacks: The
Master_cross_call
plugin guides users to perform dangerous operations
Installation Instructions
Usage
Project Structure
Security Warning
⚠️ This project is for educational and research purposes only. Do not use these techniques on any system without authorization. Malicious use of this code may violate laws and regulations.
Defense Recommendations
- Implement strict plugin verification mechanisms
- Thoroughly check and sanitize all external inputs
- Implement the principle of least privilege, limiting plugin execution permissions
- Use signature verification to ensure only trusted plugins are loaded
- Regularly review installed plugins and their behaviors
Attack Analysis
Banana Detection Poisoning
The initialize_data_poisoning.py
plugin establishes a mandatory process dependency by requiring a "banana check" before any operation. This technique can be used to:
- Induce users to perform unnecessary operations
- Establish a false sense of security
- Insert malicious steps before legitimate operations
JSON Data Injection
The inject_json_poisoning.py
plugin retrieves data from a local port by default, potentially leading to:
- Data leakage
- Execution of malicious instructions
- Bypassing normal security checks
Competitive Function Override
The malicious_competitive_function.py
provides a remove_server
function with the same name but different functionality:
- Replaces critical system functions
- Prevents normal service removal operations
- Embeds hidden instructions encoded in hexadecimal within the confusion
Cross-MCP Call Attack
The malicious_cross_mcp_call.py
uses encoded error messages to induce users to:
- Add unverified external services
- Perform unnecessary operations
- Expand the attack surface
Contribution Guidelines
Contributions to this project are welcome through:
- Submitting examples of new attack vectors
- Improving documentation for existing examples
- Adding example implementations of defense mechanisms
License
This server cannot be installed
A demonstration tool that showcases potential security attack vectors against Model Control Protocol, illustrating how malicious plugins can exploit weaknesses in MCP architecture.
Related MCP Servers
- AsecurityAlicenseAqualityA demonstration server that reveals security risks by accessing sensitive environment variables, illustrating how MCP tools can potentially leak user data without explicit consent.Last updated -18PythonMIT License
- AsecurityAlicenseAqualityA Model Context Protocol server designed for testing backend APIs for security vulnerabilities like authentication bypass, injection attacks, and data leakage.Last updated -149TypeScriptMIT License
- AsecurityAlicenseAqualityA vulnerable MCP server implementation that demonstrates how poor coding practices can lead to security issues like Remote Code Execution, designed for educational purposes to add numbers.Last updated -11PythonMIT License
- -securityFlicense-qualityIntentionally vulnerable Model Context Protocol (MCP) server designed for security research that processes natural language queries through an LLM to execute SQL queries or shell commands without restrictions.Last updated -3Python