alterlab-adaptyv

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow and examples (SKILL.md and reference/*.md) show runtime HTTP requests to external services — e.g., the Adaptyv API base URL (https://kq5jp7…on.aws), external prediction services like NetSolP (https://services.healthtech.dtu.dk), and arbitrary result download_urls — and the agent is instructed to parse those third‑party responses to rank/select sequences and drive follow-up actions, so untrusted external content can materially influence behavior.