mlx

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

• Third-party content exposure detected (high risk: 0.90). The skill directly loads and converts models from the public Hugging Face hub (e.g., load("mlx-community/...") and commands using --hf-path) and ingests arbitrary JSONL training data for fine-tuning, which are untrusted, user-provided third‑party sources that the agent will read/use in its workflow.