Skip to main content
SAICA-KG

Tools

AI-Red-Teaming-Playground-Labs

1,915

AI red-teaming labs: turnkey training environments for attacking AI systems.

DetectionPost-generationFull HITLExperimentalTier 3Environment
Maturity tier details

Computed tier: 3(raw score 3.81). Effective tier: 3.

Facets:documented_in=0evaluated_on=0integration_surfaces=0.5paper_citations=0recent_commit=2stars=1.31

Source of truth:pipeline/maturity/score.py

Description

Microsoft's AI Red Teaming Playground Labs is a curated set of reproducible training environments for hands-on AI red-teaming education — covers prompt injection, data exfiltration, and agent-exploit scenarios. Used for workforce training and capability benchmarking.

Why it's included

Training-focused rather than probe-focused: fills the 'red-team curriculum' niche, valuable as a skills development complement to runtime scanners.

Addresses failure modes

  • Security vulnerability — Insecure patterns, injection flaws, weak cryptography, secret leakage.
  • Fabrication — Generating reference to non-existent APIs, packages, or symbols.
  • Scope creep — Actions outside declared task boundary.

Techniques

red-team-labstraining-environmentscapture-the-flag

View raw YAML on GitHub →