analyze-test-run

You can analyze GitHub Actions test runs by providing a Run ID or URL to download artifacts, parse junit.xml, and extract per-test results. This process automatically summarizes outcomes and surfaces failure root causes for integration tests like azure-deploy.

The best way to find the root cause of integration test failures is to download test run artifacts, parse junit.xml, and extract per-test results to produce a structured report. This identifies failure patterns and computes skill invocation rates to pinpoint issues.

You can compare two GitHub Actions test runs to spot regressions by computing skill invocation rates and extracting per-test results. This comparison highlights performance changes and new failures across the selected runs.

This test run analysis approach supports downloading SKILL-REPORT.md and agent-metadata-*.md files alongside junit.xml. It fetches these artifacts from GitHub Actions to generate a comprehensive structured skill report suitable for automated review.

You file GitHub issues for test failures with root-cause analysis by generating a structured skill report from downloaded artifacts and parsed junit.xml results. This report accelerates remediation by providing actionable failure insights directly in the issue.