← back to the directory
Library / SDK Guardrails & Structure

PyRIT

1Red-team 2generative 3AI 4systems 5for 6risk.

the six Ws · specification

W1 Who

Built by Microsoft's AI Red Team.

W2 What

Automates adversarial probing to surface security and safety risks in AI systems.

W3 Where

A Python framework against your endpoints.

W4 When

When assessing an AI system's risk profile.

W5 Why

Brings systematic red-teaming to generative AI.

W6 With

Python and target model access.

W7 Watch

MIT, from Microsoft; sends adversarial prompts to the systems you point it at.

MicrosoftRed-teamingSecurity

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.