alterlab-fda

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly queries public openFDA endpoints (e.g., https://api.fda.gov/* as shown in SKILL.md and in scripts/fda_query.py and references/*.md) to ingest untrusted, user‑submitted reports and web API data which the agent reads and uses to drive analyses and decisions (examples in scripts/fda_examples.py), so third‑party content can materially influence its actions.