diagnose

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

• Third-party content exposure detected (high risk: 0.70). The skill's diagnostic steps explicitly probe and rely on public Hugging Face resources (references/common-issues.md: "Test HuggingFace connectivity" using curl https://huggingface.co/ and checks of ~/.cache/huggingface/hub/models--.../ plus kokoro-install.sh downloads), so it ingests untrusted third‑party content that can influence troubleshooting actions.