bio-admet-prediction

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs calling the public ADMETlab 3.0 REST API (https://admetlab3.scbdd.com/api/predict) in SKILL.md, usage-guide.md, and examples/predict_admet.py, ingesting and parsing its JSON predictions to drive filtering and prioritization decisions, which exposes the agent to untrusted third‑party content that can influence actions.