engineering-backend-architect

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

• Third-party content exposure detected (high risk: 0.70). The skill includes a health-check script (scripts/check_api_health.sh) that fetches arbitrary base URLs and prints/truncates response bodies (and the SKILL.md lists this script as part of its workflow), so the agent would ingest untrusted third‑party HTTP content that could influence its next actions.