modal

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

• Third-party content exposure detected (high risk: 0.80). The skill clearly fetches and ingests untrusted third‑party content — e.g., SKILL.md and references/images.md show running git clone from a public GitHub repo and downloading models via AutoModel.from_pretrained (Hugging Face), and references/examples.md shows requests.get to external APIs — and those fetched artifacts (code, models, API responses) are executed/used at runtime and can therefore materially alter tool behavior.