biobankjapan-phewas-skill

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

• Third-party content exposure detected (high risk: 0.70). The skill's runtime explicitly fetches variant data from public third-party endpoints (scripts/biobankjapan_phewas.py calls https://pheweb.jp/api/variant/... and scripts/variant_resolution.py calls Ensembl REST APIs at https://rest.ensembl.org and https://grch37.rest.ensembl.org), and that untrusted external JSON is parsed and used to decide outputs and warnings as part of the workflow, so third-party content could influence the agent's behavior.