npu-torchair-infer

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.85). The codebase is not overtly malicious but intentionally includes high-risk features: it downloads model artifacts from configurable external endpoints and exposes AutoModel.from_pretrained(..., trust_remote_code=True), which permits execution of arbitrary remote repository code and thus enables remote code execution / supply-chain attacks if untrusted models or endpoints are used.