regression-testing

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow triggers GitHub Actions runs via the GitHub CLI, downloads CI artifacts (Allure reports, Playwright reports, and failure evidence including test logs, traces, and error messages), and ingests failure outputs for classification.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill contains a reference URL pointing to an external GitHub raw file (https://raw.githubusercontent.com/upex-galaxy/agentic-qa-boilerplate/main/allurerc.mjs) used during a runtime migration flow, representing an unverifiable remote external dependency of weak/unknown provenance.