Popular repositories Loading

1. quickset quickset Public

   A reproducible benchmark for Machine Learning model-file security scanners

   Python 1 1

2. hayward hayward Public

   Security scanner for machine-learning model files: detects code execution in pickle, GGUF, ONNX, Keras and other checkpoints

   Python 1 1

3. langfail langfail Public

   Deliberately vulnerable Flask/MLOps app: benchmark target for taint analysis and agentic vulnerability discovery. Education and research only.

   Python 1

4. Real-Vuln-Benchmark Real-Vuln-Benchmark Public

   Forked from kolega-ai/Real-Vuln-Benchmark

   A benchmark for SAST security scanners against a suite of realistic OS apps.

   HTML

5. billhook billhook Public

   An independent harness for comparing prompt-injection defenses at a matched false-positive rate.

   Python 1

6. langchain-hardening-rules langchain-hardening-rules Public

   Open LangChain rule pack from Rowan, Hedgerow's SAST tool. OpenGrep/Semgrep rules that flag dangerous LangChain usage. MIT.

   Python 1