The Eval Index / Benchmarks / #271
prime-radiant-inc/scenarios
by prime-radiant-inc · Benchmarks · updated 4mo ago
A portable, model-agnostic test suite of declarative YAML scenarios for evaluating AI assistant security against prompt-injection and data-exfiltration attacks, with Python tools and pluggable harnesses.
18
momentum
11
stars
2
forks
#271
rank
ai-securityevalsllmmcpprompt-injectionpython
View on GitHub →