Skip to content
@q-uestionable-AI

q-uestionable-AI

Security research tools for agentic AI infrastructure.

MCP server scanning, traffic interception, tool poisoning, agent-chain exploitation, indirect prompt injection, context poisoning, retrieval poisoning


The Platform

{q-AI} is a unified offensive security platform with seven research modules for testing agentic AI infrastructure end-to-end.

Module Focus
audit Automated MCP server scanning mapped to the OWASP MCP Top 10
proxy Interactive interception proxy for MCP traffic
inject Tool-output poisoning and prompt injection testing against any LLM provider
chain Multi-step attack chain execution across trust boundaries
ipi Indirect prompt injection across 7 document formats with callback tracking
cxp Coding assistant context-file poisoning across 6 IDE formats
rxp Retrieval-layer adversarial measurement for RAG poisoning

Repository: https://github.com/q-uestionable-AI/qai Documentation: https://docs.q-uestionable.ai


Install

pip install q-uestionable-ai

Research


Responsible Use

These are offensive security testing tools. Only test systems you own, control, or have explicit permission to test.

Popular repositories Loading

  1. .github .github Public

  2. qai qai Public

    Offensive security platform for agentic AI infrastructure

    Python

Repositories

Showing 2 of 2 repositories

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…