deployment-verification

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow and scripts (e.g., SKILL.md examples and scripts/verify-vercel.sh and scripts/verify-gcp.sh) explicitly fetch and parse content from arbitrary deployment URLs and health endpoints (e.g., curl https://your-app.vercel.app/api/health, vercel ls/inspect, gcloud run services describe), meaning untrusted third-party pages can be read and their responses used to decide checks and next actions, enabling indirect prompt-injection risk.