Repo

injectbench

@3nesdenizby Enes Deniz · AltaySec

0

Reproducible benchmark for prompt-injection detectors & LLM guardrails: paired benign/attack boundary metrics, false-positive-aware scoring, per-family/per-language breakdowns, SARIF output.

Built withPythonai-securitybenchmarkblue-teamguardrailsllm

See what other members are building and what each project is looking for.