Where are the downloaded CI artifacts stored?

Artifacts are stored locally in the '.work/prow-job-analyze-test-failure/' directory, which is typically excluded from version control via .gitignore.

How does it identify the cause of a test failure?

It parses the build-log.txt for stack traces, inspects E2E timeline interval files for 'OperatorState' errors, and cross-references these with your local source code.

Does this skill require external tools?

Yes, it requires the gcloud CLI to be installed and authenticated to access the Google Cloud Storage buckets where artifacts are stored.

Can I re-run an analysis on the same job?

Yes, the skill detects existing artifacts and asks whether you want to use the local cache or re-download the latest data from the CI job.

What kind of Prow Job URLs are supported?

The skill supports gcsweb URLs containing 'test-platform-results' that point to Prow CI logs and metadata.

Prow Job Test Failure Analyzer

Name: Prow Job Test Failure Analyzer
Author: openshift-eng

byopenshift-eng

•

セキュリティとテスト

Streamlines CI/CD debugging by automatically analyzing Prow test failures through log inspection, artifact scanning, and cluster event correlation.

This skill empowers developers to quickly diagnose failed CI tests in Prow environments by automating the tedious process of log retrieval and analysis. It downloads critical artifacts like build-log.txt and prowjob.json, parses stack traces, and cross-references them with cluster timelines and operator events to identify root causes. By correlating test failure timestamps with cluster-wide errors or warnings, the skill provides a comprehensive view of whether a failure was caused by code bugs or infrastructure instability, significantly reducing the Mean Time to Repair (MTTR) for complex E2E test suites.

主な機能

01Intelligent stack trace extraction and specific test failure parsing.

0229 GitHub stars

03Cluster timeline analysis to correlate failures with operator-level events.

04Local caching and management of CI artifacts for efficient debugging.

05Automated retrieval of Prow CI artifacts and build logs from GCS storage.

06Context-aware root cause analysis linking test logs to project source code.

ユースケース

01Rapidly debugging flaky E2E tests in Kubernetes or OpenShift environments.

02Correlating specific test failures with underlying cluster stability or operator errors.

03Extracting and organizing artifacts from remote Prow jobs for offline forensic analysis.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add openshift-eng/ai-helpers prow-job-analyze-test-failure

For use in Claude.ai and ChatGPT

Download Skill