Skip to main content

Tools

The Python Risk Identification Tool for generative AI (PyRIT).

Detection Post-generation Full HITL AI-based Experimental Activity band D Environment
Activity band details

Activity band A to F: a bin of a score that adds GitHub stars (log scale), a push to the upstream repository in the 12 months before `as_of`, the number of declared integration surfaces and, where the catalogue records them, citing papers and benchmark evaluations. It measures popularity and activity. It is not a level of evidence: no band means a paper cites the tool or that it was measured. Recency is judged as of 2026-11-22.

Computed band: D (raw score 4.48). Shown band: D .

Facets: documented_in=0evaluated_on=0integration_surfaces=1paper_citations=0recent_push=2stars=1.48

Source of truth: pipeline/maturity/score.py

Description

Microsoft PyRIT is an open-source framework for AI red-teaming. Automates probing of generative AI systems for risks (prompt injection, harmful content, jailbreaks) via a plug-in orchestrator with configurable attack and scoring strategies.

Why it's included

Microsoft-ecosystem red-team framework — complements garak with an orchestration pattern that maps well onto azure-hosted workloads.

Addresses failure modes

These tags are the catalog's claim about what PyRIT is built to catch or prevent. A tag is not a test result. SAICA-Bench has not measured PyRIT, so none of these tags is tested here.

Techniques

red-team-orchestratorattack-strategiesscoring-engines

View raw YAML on GitHub →