Downloads · 30 days
0
ryansecuritytest-fanpierlabs/onnxruntime-tensorrt-rtx-path-traversal-poc
onnxruntime-tensorrt-rtx-path-traversal-poc is a machine learning model from ryansecuritytest-fanpierlabs. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
This repository contains a proof-of-concept for a security vulnerability. It is intended for responsible disclosure via huntr.com.
Downloads · 30 days
0
Access
Public
Updated Mar 30, 2026
Repo size
349 B
Likes
0
Public
Click a slice to open those files.
.py11.7 KB · 80%
From the Hugging Face model README
This repository contains a proof-of-concept for a security vulnerability. It is intended for responsible disclosure via huntr.com.
Inverted boolean logic in onnx_ctx_model_helper.cc (line 321) of the NV TensorRT RTX execution provider disables path traversal security checks when loading EP context models from file paths.
Vulnerable code:
bool make_secure_path_checks = ep_context_model_path_.empty();
// Should be: !ep_context_model_path_.empty();
malicious_ep_context_model.onnx — Crafted ONNX model with path traversal payload in onnx_model_filename attributepoc_onnxruntime_tensorrt_rtx.py — Script that generates the PoC model and explains the vulnerabilityArbitrary file read when loading attacker-supplied ONNX models with the NvTensorRTRTX execution provider.
Ryan — Fan Pier Labs ([email protected])