dataverse-api

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly calls the public Harvard Dataverse API (e.g., https://dataverse.harvard.edu/api search and /access/datafile endpoints shown in SKILL.md) to fetch dataset metadata and user-uploaded files, which are untrusted third-party contents that the agent reads and uses to decide which results/files to present or download.