What problem does it solve? Intermittent test failures erode trust in CI pipelines and waste engineering time on re-runs. This Skill provides a structured forensic workflow to identify, reproduce, root-cause, and permanently fix flaky tests instead of masking them with retries or skips. ## Core Features & Use Cases - 9-Step Forensic Workflow: Guides you from identifying the flake in CI, downloading failure logs with the GitHub CLI, classifying the failure, reproducing it locally, through to fixing, verifying, and documenting. - 20-Category Flakiness Taxonomy: Maps error signals (timeouts, EADDRINUSE, ordering failures, timezone drift) to root-cause categories and their common fixes. - Sibling Sweep: Searches the test suite for other instances of the same antipattern (raw setTimeout, unrestored env vars, missing afterEach cleanup) so the fix is applied everywhere. - Use Case: A test passes locally but fails intermittently in CI. The Skill pulls the failed run logs, classifies the failure as a shared-state issue, reproduces it with repeated runs, fixes the root cause, and verifies with 20 consecutive passing runs. ## Quick Start Ask the agent to investigate and fix the intermittently failing test in your CI pipeline using the flaky test workflow.
