nyc-hpd

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md workflow explicitly instructs the agent to fetch and ingest open/public Socrata datasets from data.cityofnewyork.us (PLUTO 64uk-42ks and HPD datasets wvxf-dwi5, csn4-vhvf, ygpa-z7cr, tesw-yqqr in Steps 2–3), which includes public complaint/description text that the agent reads and uses to make decisions (e.g., flagging hazards), so untrusted third‑party content could influence behavior.