staging-validation-phase

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly fetches and parses staging web pages and API endpoints as part of its workflow (e.g., "npm run test:smoke -- --url=$STAGING_URL", curl/psql/health checks, and rollback verification that inspects response headers in SKILL.md and references/reference.md) and uses those responses to make validation and sign-off decisions, so untrusted third‑party page/content could materially influence actions.