crawlora-datasets

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow sends user-provided free-text queries to Crawlora search endpoints (e.g., /datasets/<id>/search with q/full-text query) and then ingests the resulting dataset records from Crawlora’s search index (which can include outsider-authored text such as reviews/news/social posts), so an outsider can poison by submitting crafted content to those underlying data sources that the search index returns.