dataset-finder-guide

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md includes concrete workflow and code that fetches and ingests public, user-generated content (e.g., the Zenodo API search_zenodo example, Hugging Face list_datasets/load_dataset snippets, and Kaggle CLI download instructions) from open repositories, which the agent would read/interpret and could materially influence subsequent decisions or tool use, enabling indirect prompt injection.