parity-testing

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs running tools that load arbitrary Hugging Face model IDs/paths (e.g., "--hf-model-id /" and "--hf_model_path /" in the examples/conversion commands) which fetch and run untrusted, user-uploaded models from the public HF hub and therefore can influence generation/verification behavior.