Benchmarks tool poisoning attacks on Minecraft Companion Protocol servers to identify vulnerabilities.
npx skills add https://github.com/zzw4257/security-skills --skill mcptox
Please help me install this Agent Skill. Skill: MCPTox Source: https://github.com/zzw4257/security-skills/tree/main/skills/mcptox Command: npx skills add https://github.com/zzw4257/security-skills --skill mcptox
This Skill addresses the challenge of evaluating the security of MCP (Minecraft Companion Protocol) servers against tool poisoning attacks, providing a benchmark for assessing vulnerabilities.
Run the MCPTox benchmark against the target MCP server.
High-intent search queries and answers about installing and using this skill.
An MCP server tool poisoning attack exploits vulnerabilities in Minecraft Companion Protocol implementations by injecting malicious inputs. This Skill benchmarks these attacks to identify security weaknesses and assess server susceptibility to known poisoning techniques.
You benchmark MCP server security vulnerabilities by running the MCPTox benchmark against the target server. This process tests the implementation to identify potential weaknesses and compare its security posture against known tool poisoning techniques.
Testing Minecraft Companion Protocol implementations requires Python 3.10 or higher and specific libraries for execution. These dependencies are necessary to run the vulnerability assessment and security benchmarking scripts against the target server.
Yes, you can use this for security auditing before a Minecraft server release. Security researchers can test newly developed MCP servers for susceptibility to known poisoning techniques to ensure vulnerabilities are patched prior to public deployment.
A standardized MCP security benchmark provides a specialized method to test and compare tool poisoning vulnerabilities specific to Minecraft Companion Protocol servers. Unlike general penetration testing, it focuses on known poisoning techniques to provide consistent vulnerability assessments.
shiromacConfigure the WPF UI automation agent and register its MCP server.
YukiIto1999Identify and remediate AI security vulnerabilities in LLM applications.
prince3626ezechiel-langMonitor Hermes runtime traffic for exfiltration and injection payloads.
RheasilviaPerform authorized web application penetration testing with reconnaissance, vulnerability analysis, and proof-based exploitation.
