analyze-test-run

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly downloads and parses GitHub Actions run artifacts (via "gh run download --repo microsoft/GitHub-Copilot-for-Azure") — including SKILL-REPORT.md and agent-metadata-*.md — and uses their untrusted, user-generated contents to determine root causes and drive issue-creation, meaning third-party artifact content can influence agent decisions and tool use.