Analyze Test Run helps an agent process GitHub Actions integration test runs by downloading artifacts, parsing test results, and calculating skill invocation rates. It builds a structured markdown report detailing test pass rates, confidence scores, and comparison metrics across runs. Additionally, it diagnoses test failures, categorizes root causes, and files corresponding GitHub issues using MCP tools. Reach for it when you need to automate test run evaluation and triage failures for GitHub-Copilot-for-Azure.