authoritative-data-harvester

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

• Third-party content exposure detected (high risk: 0.90). The skill's main workflow and script clearly fetch and ingest open web content (scripts/run.py uses requests.get and pandas.read_html to download arbitrary URLs—e.g., the en.wikipedia.org template—and SKILL.md explicitly directs fetching public data sources like World Bank/Wikipedia/statistics sites), so the agent will consume untrusted third-party content that can influence its data-processing decisions.