Demonstrates MCP Poisoning Attacks affecting AI agent workflows.
Sponsored
This repository provides a proof-of-concept demonstration of various MCP Poisoning Attacks that can compromise real-world AI agent workflows. It showcases scenarios like code generation poisoning, financial report exfiltration, competitor analysis data leaks, meeting transcript leaks, code review exfiltration, and cross-server shadowing attacks. The PoC includes a fake MCP server and an agent simulation to illustrate the impact of these attacks, which can lead to silent data exfiltration and cross-tool hijacking without visible clues to the user. This is intended for educational and research purposes only.
主な機能
01Simulates financial report exfiltration
02Shows code review exfiltration
03Illustrates competitor analysis data leaks
04Implements a cross-server shadowing attack
05Demonstrates code generation poisoning
06Replicates meeting transcript leaks
074 GitHub stars
ユースケース
01Researching AI security vulnerabilities
02Testing the security of AI agent workflows
03Educating developers about MCP poisoning attacks
