wa:validate-ci

To diagnose CI pipeline failures, this skill collects failing run logs, extracts the failing step and error evidence, and classifies the root cause. It provides evidence-backed classifications with confidence levels to explain why your GitHub Actions or GitLab CI build failed.

Common root cause classifications for CI test failures include Code Bug, Test Bug, Config, Dependency, Flaky, and Infra issues. This classification mechanism helps identify whether your CI failure stems from code changes, test logic, configuration, dependency versioning, flaky behavior, or runner infrastructure.

Yes, you can get a proposed fix for a failing CI run. After classifying the root cause of your GitHub Actions or GitLab CI failure, the skill generates a targeted, smallest-safe fix proposal complete with validation steps to help you resolve the issue.

Yes, this CI triage process works with specific run identifiers across different branches. It applies to GitHub Actions, GitLab CI, and other CI systems specified by project constants, allowing you to target and diagnose individual pipeline runs by their unique identifiers.

The best way to triage flaky tests in a CI pipeline is to analyze failing run logs and classify the root cause. This skill identifies flaky behavior by extracting error evidence and assigning confidence levels, distinguishing genuine flakiness from code or infrastructure failures.

To validate a proposed fix for a configuration or dependency error in CI, this skill produces a targeted, smallest-safe fix and a validation command. These validation steps are tailored to the specific root cause classification extracted from your CI logs.